Skip to main content

Full text of "Insect Repellents Principles, Methods, And Uses"

See other formats


Principles. Mods, and Uses 

















INSECT REPELLENTS 

Principles, Methods, and Uses 


© 2006 by Taylor & Francis Group, LLC 



© 2006 by Taylor & Francis Group, LLC 



INSECT REPELLENTS 

Principles, Methods, and Uses 


Musfapho Debboun. Phi, BCE 

U.S. flrmij Medical Department Center & School 
Fort Sam Houston, Texas, U.S.R. 

Stephen P. Frances. Ph.D. 

Rustralian Hrmij Malaria Institute 
Enoggera, Queensland, Australia 

Daniel Sfrickman, Ph.D. 

USD! Agricultural Research Service 
Beltsville, Margland, U.S.A. 



CRC Press 

Taylor St Francis Group 
Boca Raton London New York 


© 2006 by Taylor & Francis Group, LLC 


CRC Press is an imprint of the 

Taylor & Francis Group, an informa business 



Front cover: AgamOBPl — structural diagram of an odorant binding protein of the malaria-transmitting mosquito, 
Anopheles gambiae. Prepared by and used with permission from Professor Walter S. Leal, University of California, 
Davis. 


CRC Press 

Taylor & Francis Group 

6000 Broken Sound Parkway NW, Suite 300 

Boca Raton, FL 33487-2742 

© 2007 by Taylor & Francis Group, LLC 

CRC Press is an imprint of Taylor & Francis Group, an Informa business 

No claim to original U.S. Government works 

Printed in the United States of America on acid-free paper 

10 987654321 

International Standard Book Number-10: 0-8493-7196-1 (Hardcover) 

International Standard Book Number-13: 978-0-8493-7196-7 (Hardcover) 

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted 
with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to 
publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of 
all materials or for the consequences of their use. 

No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or 
other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any informa¬ 
tion storage or retrieval system, without written permission from the publishers. 

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http:// 
www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 
978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For orga¬ 
nizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. 

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for 
identification and explanation without intent to infringe. 


Library of Congress Cataloging-in-Publication Data 


Insect repellents : principles, methods, and uses / Mustapha Debboun, Stephen P. Frances, and Daniel 
Strickman, editors, 
p. cm. 

Includes bibliographical references and index. 

ISBN 0-8493-7196-1 (978-0-8493-7196-7) 

1. Insect baits and repellents. I. Debboun, Mustapha. II. Frances, Stephen P. III. Strickman, Daniel. 
SB951.I47 2006 

614.4'32—dc22 2006012882 


Visit the Taylor & Francis Web site at 
http://www.taylorandfrancis.com 

and the CRC Press Web site at 
http://www.crcpress.com 


© 2006 by Taylor & Francis Group, LLC 



Dedicated to 


My dear four brothers and four sisters; my beautiful and loving wife, Natalie, and our wonderful 
children: Ameena, Adam and David; and to the memory of my beloved parents 

Philip Granett, Rutgers University, New Brunszvick, New jersey, USA 

Carroll N. Smith, Harry K. Gouck, T. P. McGovern, and Carl E. Schreck, 
Agricultural Research Service, U.S. Department of Agriculture; Orlando, 

Florida; Beltsville, Maryland; and Gainesville, Florida, USA 

Robert N. McCidloch and Douglas F. Waterhouse, Commonwealth Scientific 
and Industrial Research Organisation, Canberra, Australian Capital Territory, Australia 

Pioneers, leaders, intellects, and good friends of insect repellent science 


© 2006 by Taylor & Francis Group, LLC 



© 2006 by Taylor & Francis Group, LLC 



Preface 


The use of repellent products to prevent insect and arthropod bites is probably proportional to the public 
perception of the threat from biting arthropods, whether the threat is from annoyance or from the risk of 
disease. The connection between perception and use is logical when one considers that repellents are 
generally used as personal protection. It is the individual who usually decides whether or not to use a 
repellent, what kind of repellent to use, and how much to apply. The application to the individual makes 
entomological sense, in that the countermeasure is applied exactly where it is needed. On the other hand, 
the application by the individual presents a challenge to the professional who must educate and inform 
people with widely varied backgrounds on how to best protect themselves from biting arthropods. 

Insect and arthropod repellents are usually the first line of defense because they require no large 
equipment, no organized effort of community vector control, and they distribute the responsibility for 
protection to the individual. Today, there is great public concern throughout the world about vector- 
borne pathogens as human ecology seems to favor outbreaks of diseases as varied as West Nile fever, 
dengue, Lyme disease, malaria, leishmaniasis, and tick-borne encephalitis. In contrast to community 
vector control programs, insect and arthropod repellents give the individual control over exposure to 
biting arthropods. Professional researchers in public health are also interested in the development and use 
of repellents given the increasingly complicated requirements for pesticide use, the high cost of 
developing effective prophylactic vaccines and drugs, and the increase in incidences of arthropod- 
borne diseases. 

All of the insect and arthropod repellent literature is in the form of individual articles, reviews, 
symposia, commercial literature, book chapters in medical entomology texts, etc. We are only aware of 
two volumes dedicated solely to repellents, and those were handbooks published by the U.S. Department 
of Agriculture (USDA) in the 1950s. Our objective was to provide a one-volume source for most aspects 
of the development and use of repellent products designed to protect people from biting arthropods. As 
the title implies, parts of this book cover the theory and science (principles), the means for advancing the 
particular area (use of standard methods for future product development and testing), and the 
implications for effective protection of people from biting arthropods. 

Although most of the writing is technical, the informed public, physicians, public health officials, and 
other nonspecialists will find this book easily comprehensible. We hope that the following groups will get 
specific benefit from the book: The public will be able to choose the proper repellent product for their 
situation and use it more effectively. Medical professionals will be able to make better recommendations 
to patients who are seeking safe and effective means of preventing arthropod bites and arthropod-borne 
diseases in particular situations. Public health personnel will be able to integrate repellents more 
effectively into programs to limit arthropod-borne diseases and better inform travelers about protection in 
unfamiliar parts of the world. Medical entomologists will be able to perform evaluations with greater 
knowledge of theoretical concerns and using more standard techniques. Hopefully, the two appendices 
will be useful to those who want to identify common chemicals and organisms used in this line of 
research. 

While editing this book, we found that the very word “repellent” was used with many different 
meanings. Although there is a rich vocabulary of terms for chemicals affecting arthropod behavior, none 
combined the specificity about the feeding process while retaining the general application to the many 
different aspects of that process. With some reluctance, we introduce the term "phagomone” to fill this 
gap and we suggest that the term “repellent” be restricted to products, rather than to any technical 
description of a particular chemical that affects behavior. Phagomone designates any chemical that 
influences the arthropod feeding process, whether on the side of facilitating feeding (attractants, 
phagostimulants, host-recognition factors, etc.) or on the side of discouraging feeding (irritants. 


© 2006 by Taylor & Francis Group, LLC 



disorienting toxicants, blockers of receptors, compounds with gustatory effect or with olfactory effect, 
etc.). People have been creative in the use of phagomones but much less active in performing the science 
to understand how these chemicals work. Use of the term should allow researchers to discuss their work 
more accurately, particularly when they are at the stage when they do not completely understand the 
exact aspect of feeding behavior disrupted or enhanced by the phagomone. 

The first part of the book treats some of the basic principles behind the use of repellent products. 
Starting with a history of repellent product development that has led to the formulations in use today and 
proposed for the future, this part proceeds to a discussion of terminology that attempts to specify what has 
been a confusing vocabulary used in the field. Some of the biological variety of phagomones as they 
function in nature are presented in two chapters on naturally occurring compounds in vertebrates, 
including humans. 

The chapters in the second part address the methods used to assess the activity of phagomones and 
repellent products. The authors of these chapters present many careers’ worth of experience in this field, 
but the reader will soon see that the experts’ opinions do not always correspond to each other. What seem 
like standard methods to one author might lack sufficient statistical rigor for another. Whereas one school 
of thought might advocate the use of human trials in the field, another advocates the use of animal models 
under controlled conditions. The reader should be able to gain a good appreciation of the variety of 
purposes served by repellent bioassays and then be better prepared to evaluate data and design new tests. 
Apart from traditional bioassays of biting behavior, this part also presents what may be the cutting edge 
of repellent testing: automated tests of the arthropods’ responses, the use of computer models of 
chemistry, and the use of molecular biology methods. 

The third part of the book concentrates on repellent products that have come to market at one time or 
another. Following a thoughtful discussion of the process of testing formulated products, the part 
includes comprehensive reviews of both natural and synthetic active ingredients. Currently, the most 
important active ingredients are deet, Picaridin, PMD. DEPA, and 1R3535, but older active ingredients 
were very useful in their time. Some of the older active ingredients are still used as additives that have a 
somewhat synergistic effect on product performance. The botanical active ingredients are very 
interesting in their variety and origins, illustrating Gene Gerberg’s wise comment that when it comes 
to repellents, “one size does not fit all.” 

The fourth and final part of the book deals with the concept of the use of insect repellents. User 
acceptability and public perceptions of insect repellents are discussed at length because a repellent 
product cannot perform well if it is not used. The great variety of commercially available products shows 
how the business community continues to try to perfect its efforts and satisfy what the public considers its 
needs. The review of global regulatory procedures is just an introduction to this process, but it 
emphasizes the supreme importance of regulation of products that are applied directly to the skin. We 
conclude the volume with an epilogue in which we indulge ourselves in some speculation on where the 
fields of phagomone and repellent product research have been and where they are going. 

As the editors of this volume, we thank the authors and their employers, who generously donated their 
time and centuries of accumulated professional experience. We also thank Jill Jurgensen and 
John Sulzycki of CRC/Taylor & Francis, who patiently guided us through the process of translating 
an idea into a book. 


Mustapha Debboun, San Antonio, Texas, USA 
Stephen P. Frances, Enoggera, Queensland, Australia 
Daniel Strickman, Beltsville, Maryland, USA 


© 2006 by Taylor & Francis Group, LLC 



Acknowledgments 


Major (Dr.) Frances would like to thank Professor Karl Rieckmann (former Director) and Lieutenant 
Colonel (Dr.) Robert Cooper (Commanding Officer), Australian Army Malaria Institute, for their support 
over many years. The opinions expressed in his chapters are his alone, and do not reflect those of the 
Australian Defence Health Service or any extant defence policy. 


© 2006 by Taylor & Francis Group, LLC 



© 2006 by Taylor & Francis Group, LLC 



Editors 


Dr. Mustapha Debboun is a medical and veterinary entomologist in the 
United States (U.S.) Army Medical Department. He was born in Tangier, 
Morocco and commissioned as an officer in the U.S. Army Medical 
Service Corps in 1989 after receiving his doctoral degree. He received 
his B.A. degree in cellular and molecular biology from Skidmore College 
in Saratoga Springs, New York, a master of science degree in medical 
entomology from the University of New Hampshire, Durham, New 
Hampshire, and a doctor of philosophy degree in medical and veterinary 
entomology from the University of Missouri-Columbia. 

Dr. Debboun joined the U.S. Army in 1989 and served in assignments at 
the 44th Medical Brigade, Fort Bragg, North Carolina, Operations “Desert Shield" and “Desert Storm,” 
Saudi Arabia; Academy of Health Sciences, U.S. Army Medical Department Center and School, Fort 
Sam Houston, Texas; U.S. Army Europe, Commander, 255th Medical Detachment, Vicenza, Italy, 
“Operation Joint Endeavor,” Bosnia; Office of the Assistant Secretary of the Army for Research, 
Development, and Acquisition, Pentagon, Arlington, Virginia; Walter Reed Army Institute of Research, 
Washington, D.C.; 3rd Medical Command, Operations “Enduring Freedom” and "Iraqi Freedom,” Camp 
Arifjan, Kuwait and Camp Bucca, Iraq; and Commander, U.S. Army Center for Health Promotion and 
Preventive Medicine-South, Fort McPherson, Georgia. 

During 17 years in the military. Dr. Debboun has worked in preventive medicine operations, research 
and development of arthropod repellents, and field personal protective measures. This work has taken 
him to 20 different countries in Africa, Asia, Australia, Europe, and Latin America. He is now deputy 
chief of the Department of Preventive Health Services and chief of the Medical Zoology Branch at the 
Academy of Health Sciences. He is a board certified medical and veterinary entomologist, and a member 
of the Entomological Society of America, American Mosquito Control Association, and the Society of 
Vector Ecology. He serves as adjunct faculty of the Non-Resident Command and General Staff College, 
and as chair of the Repellents Committee and vice-chair of the Education and Training Committee, 
Armed Forces Pest Management Board, Silver Spring, Maryland. Dr. Debboun has authored or co¬ 
authored more than 50 research publications. 



Dr. Stephen P. Frances is a medical entomologist at the Australian Army 
Malaria Institute, Royal Australian Army Medical Corps, in Brisbane, 
Queensland, Australia. He was born in Sydney, Australia, and attended the 
University of Sydney, graduating with a bachelor of science in agriculture 
with honors in 1980. From 1981 to 1984 he worked at the Commonwealth 
Institute of Health, Sydney, as a research assistant conducting studies on a 
fungal pathogen of mosquito larvae as a biocontrol agent under the 
supervision of Professor Richard Russell. He completed a master of 
science in agriculture in 1985. 

In 1985 he was commissioned as an officer in the Australian Defence 
Force in the Army Malaria Research Unit, Ingleburn, New South Wales, and commenced work on 
evaluating insect repellent formulations and other physical and chemical barriers to protect soldiers 
against medically important arthropods, especially mosquitoes and mites. From 1992 to 1994 he served 
in the Department of Medical Entomology, Armed Forces Research Institute of Medical Sciences, 
Bangkok, Thailand, continuing his work evaluating repellents. He commenced PhD studies in 1991 on a 
part time basis under the supervision of Dr. (Lieutenant Colonel) Anthony (Tony) Sweeney, and was 





© 2006 by Taylor & Francis Group, LLC 


awarded his doctor of philosophy from the Faculty of Medicine at the University of Sydney in 1999. His 
dissertation was written on aspects of the transmission of the rickettsia that causes scrub typhus in two 
regions of Thailand. He has also been involved in vector surveillance in northern Australia, Papua New 
Guinea, and Timor Leste. He is a member of the American Mosquito Control Association, the Mosquito 
Control Association of Australia, the Australian Entomological Society, and the Entomological Society 
of Queensland. Dr. Frances has authored or co-authored 60 research publications. 

Dr. Daniel Strickman attended Dartmouth College from 1971 to 1973, 
transferring to the University of California at Riverside to pursue his interest 
in entomology. He received his bachelor of arts degree in biology from UCR 
in 1974 and began graduate studies at the University of Illinois, Champaign- 
Urbana, in the same year. Studying under the late Dr. William Horsfall, he 
worked on oviposition habits of Midwestern mosquitoes, completing his 
master of science degree in 1976 and his doctor of philosophy degree in 1978. 
Dr. Strickman and his wife, Linda, joined the Peace Corps and served in 
Paraguay on the staff of the National University of Asuncion for two years. 
During that time, they taught environmental education and field entomology, 
and completed natural history studies of mosquitoes, horse flies, and dragonflies. Dr. Strickman joined 
the U.S. Air Force in 1981 and served as a consultant on toxicological issues throughout the United 
States. He transferred to the U.S. Army in 1984, completing assignments at the Smithsonian Institution, 
the Armed Forces Research Institute of Medical Science in Bangkok, the 5th Preventive Medicine Unit in 
Seoul, and Walter Reed Army Institute of Research in Washington, D.C. During 22 years in the military. 
Dr. Strickman worked in operations and research dealing with toxicology, taxonomy, repellents, 
rickettsial diseases, dengue, malaria, and insect control. This work took him to nine different countries 
in Latin America, Africa, and Asia. Following his retirement from the military in 2003, Dr. Strickman 
worked as the vector ecologist for the Santa Clara County (California) Vector Control District. He is now 
national program leader for veterinary, medical, and urban entomology for the Agricultural Research 
Service, U.S. Department of Agriculture. Dr. Strickman is a member of the Entomological Society of 
America, the American Mosquito Control Association, the American Society of Tropical Medicine and 
Hygiene, the National Cattlemen’s Beef Association, and the U.S. Animal Health Association. He is an 
author of 90 scientific publications and his main research interest is the integration of entomology with 
other operational fields to provide efficient, sustainable management of disease to protect humans and 
animals. 



© 2006 by Taylor & Francis Group, LLC 



Contributors 


Arshad Ali, Ph.D. 

University of Florida, IFAS 
Department of Entomology and 
Nematology 

Mid-Florida Research Education Center 

2725 Binion Road 

Apopka, FL 32703 

Phone: (407) 884-2034 

FAX: (407) 814-6186 

E-mail: aali@mail.ifas.ufl.edu 

Donald R. Barnard, Ph.D. 

USDA/ARS 

Center for Medical, Agricultural and 
Veterinary Entomology 
P.O. Box 14565 
Gainesville, FL 32604 
Phone: (352) 374-5930 
FAX: (352) 374-5870 

E-mail: dbarnard@gainesville.usda.ufl.edu 

Ulrich R. Bernier, Ph.D. 

USDA/ARS/CMAVE 
Mosquito and Fly Research Unit 
1600 SW 23rd Drive 
Gainesville, FL 32608 
Phone: (352) 374-5917 
FAX: (352) 374-5922 

E-mail: ubernier@gainesville.usda.ufl.edu 

Apurba K. Bhattacharjee, Ph.D. 

Walter Reed Army Institute of Research 

Department of Medicinal Chemistry 

Division of Experimental Theraputics 

Silver Spring, MD 20910 

Phone: (301) 319-9043 

E-mail: apurba.bhattacharjee@na.amedd. 

army.mil 

Jerry F. Butler, Ph.D. 

University of Florida 
Medical-Veterinary Entomology 
Entomology and Nematology Department 
Institute of Food and Agricultural Sciences 


Bldg. 970, Natural Area Drive 
P.O. Box 110620 
Gainesville, FL 32611-0620 
Phone: (352) 392-1930 ext. 152 
FAX: (352) 392-0190 
E-mail: jfb@ifas.ufl.edu 

John F. Carroll, Ph.D. 

USDA/ARS 

Animal Parasitic Diseases Laboratory 

BARC-East, Bldg. 1040 

Beltsville, MD 20705 

Phone: (301) 504-9017 

FAX: (301) 504-5306 

E-mail: jcarroll@anri.barc.usda.gov 

Scott P. Carroll, Ph.D. 

University of California-Davis 
Department of Entomology 
Center for Population Biology 
Davis, CA 95616 
E-mail: spcarroll@ucdavis.edu 

Jonathan F. Day, Ph.D. 

University of Florida, IFAS 

Florida Medical Entomology Laboratory 

2009th Street, SE 

Vero Beach, FL 32962 

Phone: (772) 778-7200 

E-mail: jfday@ifas.ufl.edu 

LTC Mustapha Debboun, Ph.D., BCE 

U.S. Army Medical Department Center 
and School 

Academy of Health Sciences 
Department of Preventive 
Health Services 
Medical Zoology Branch 
3151 Scott Road, Suite 0408A 
Fort Sam Houston, TX 78234-6142 
Phone: (210) 210-7649 
FAX: (210) 210-8332 
E-mail: mustapha.debboun@us. 
army.mil 


© 2006 by Taylor & Francis Group, LLC 



David N. Durrheim, DrPH, MPHandTM, 
MBChB, FAFPHM, FACTM 

Hunter New England Population Health 
Health Protection 
Locked Bag 10 

Wallsend New South Wales 2287 
Australia 

Phone: 61-2-49246473 

FAX: 61-2-49246048 

E-mail: david.durrheim@hnehealth. 

nsw.gov.au 

Major Stephen P. Frances, Ph.D. 

Australian Army Malaria Institute 
Vector Surveillance and Control 
Weary Dunlop Drive 
Gallipoli Barracks Enoggera QLD 4051 
Australia 

Phone: 61 7 3332 4807 

FAX: 61 7 3332 4800 

E-mail: steve.frances@defence.gov.au 

Eugene J. Gerberg, Ph.D. 

University of Florida, IFAS 

Entomology and Nematology Department 

P.O. Box 110620 

Gainesville, FL 32611-0620 

Phone: (home) (352) 373-7384 

E-mail: genejg2@aol.com 

John M. Govere, Ph.D. 

World Health Organization for African 
Region (WHO/AFRO) 

Parirenyatwa Hospital 

Mazoe Street 

P.O. Box CY384 

Harare, Zimbabwe 

Phone: 263 4 253724-30 

FAX: 263 4 253731-2 

E-mail: goverej@zw.afro.who.int 

Col. Raj K. Gupta, Ph.D. 

Walter Reed Army Institute of Research 

Office of the Science Director 

503 Robert Grant Avenue 

Silver Spring, MD 20910 

Phone: (301) 619-7732 

FAX: (301) 619-2982 

E-mail: raj.gupta@amedd.army.mil 


James Ha, B.A. 

University of Illinois 
College of Medicine-Rockford 
1601 Parkview Avenue 
Rockford, IL 61107 
Phone: (815) 621-0800 
FAX: (815) 395-5887 
E-mail: jha2@uic.edu 

Yi-Xun He, Ph.D. 

University of Illinois 
College of Medicine-Rockford 
1601 Parkview Avenue 
Rockford, IL 61107 
Phone: (815) 395-5694 
FAX: (815) 395-5666 
E-mail: yixunhe@uic.edu 

Nigel Hill, Ph.D. 

London School of Hygiene and Tropical 
Medicine 

Department of Infectious and Tropical 
Diseases 

Vector Biology and Disease Control Unit 

Keppel Street 

London WC1E 7HT 

United Kingdom 

Phone: 020 7927 2646 

FAX: 020 7636 8739 

E-mail: nigel.hill@lshtm.ac.uk 

Daniel L. Kline, Ph.D. 

USDA/ARS/CMAVE 
Mosquito and Fly Research Unit 
1600 SW 23rd Drive 
Gainesville, FL 32608 
Phone: (352) 374-5917 
FAX: (352) 374-5922 

E-mail: dkline@gainesville.usda.ufl.edu 

Glenn J. Leach, Ph.D. 

U.S. Army Center for Health Promotion 
and Preventive Medicine 
Directorate of Toxicology 
Toxicity Evaluation 
5158 Blackhawk Road 
Aberdeen Proving Ground, MD 21010 
FAX: (410) 436-6710 
E-mail: glenn.leachl@us.army.mil 


© 2006 by Taylor & Francis Group, LLC 



Walter S. Leal, Ph.D. 

University of California-Davis 
Department of Entomology 
308D Briggs Hall 
One Shield Ave 
Davis, CA 95616-8584 
Phone: (530) 752-7755 
FAX: (530) 754-8682 
E-mail: wsleal@ucdavis.edu 

Annick Lenglet 

London School of Hygiene and Tropical 
Medicine 

Vector Biology and Disease Control Unit 

Keppel Street 

London WC1E 7HT 

United Kingdom 

E-mail: annick@thelenglets.com 

Wilfred C. McCain, Ph.D. 

U.S. Army Center for Health Promotion and 
Preventive Medicine 
Directorate of Toxicology 
5158 Blackhawk Road 
Aberdeen Proving Ground, MD 21010 
Phone: (410) 436-2201 
FAX: (410) 436-6710 
E-mail: wilfred. mccain@us. army, mil 

Sarah J. Moore, Ph.D. 

London School of Hygiene and Tropical 
Medicine 
Keppel Street 
London WC1E 7HT 
United Kingdom 
Phone: 020 763 68636 
E-mail: sarah.moore@sjmoore.net 

Robert J. Novak, Ph.D. 

University of Illinois Urbana-Champaign 

Illinois Natural History Survey 

Medical Entomology Program 

607 E. Peabody Drive 

Champaign, Illinois 61820 

Phone: (217) 333-1186 

FAX: (217) 333-2359 

E-mail: rjnovak@uiuc.edu 

Kenneth H. Posey 

USDA/ARS/CMAVE 
Mosquito and Fly Research Unit 


1600 SW 23rd Drive 
Gainesville, FL 32608 
Phone: (352) 374-5917 
FAX: (352) 374-5922 

E-mail: kposey@gainesville.usda.ufl.edu 

Shri Prakash, Ph.D. 

Defence Research and Development 
Establishment 
Division of Entomology 
Jhansi Road 
Gwalior-474002 
M.P., India 

E-mail: shripra2004@yahoo.co.in 

Germain Puccetti, Ph.D. 

EMD Chemicals, Inc. 

7 Skyline Drive 
Hawthorne, NY 10532 
Phone: (914) 592-4660 ext 489 
FAX: (914) 785-5889 

E-mail: gpuccetti@emdchemicals.com; 
gpl000@mail.solgel.com 

Kalyanasundaram Ramaswamy, Ph.D. 

University of Illinois 

College of Medicine-Rockford 

Department of Biomedical Sciences 

1601 Parkview Avenue 
Rockford, IL 61107 
Phone: (815) 395-5696 
FAX: (815) 395-5666 
E-mail: ramswamy@uic.edu 

Louis C. Rutledge 

United States Army, Retired 
11 Circle Way 
Mill Valley, CA 94941-3420 
Phone: (415) 388-2937 
E-mail: louiscrutledge@msn.com 

Buz Salafsky, Ph.D. 

University of Illinois 
College of Medicine-Rockford 
Department of Pharmacology 
1601 Parkview Avenue 
Rockford, IL 61107 
Phone: (815) 395-5697 
FAX: (815) 395-5887 
E-mail: buzs@uic.edu 


© 2006 by Taylor & Francis Group, LLC 



K. Sekhar, Ph.D. 

Defence Research and Development 
Establishment 
Jhansi Road 
Gwalior-474 002 
M.P., India 

Takeshi Shibuya, Ph.D. 

University of Illinois 
College of Medicine-Rockford 
Department of Pharmacology 
1601 Parkview Avenue 
Rockford, IL 61107 
Phone: (815) 395-0600 
FAX: (815) 395-5887 

Daniel Strickman, Ph.D. 

Veterinary, Medical, and Urban Entomology 

USDA/ARS, National Program Staff 

5601 Sunnyside Avenue 

Beltsville, MD 20705-5134 

Phone: (301) 504-5771 

FAX: (301) 504-5467/4725 

E-mail: daniel.strickman@ars.usda.gov 

Kevin J. Sweeney 

U.S. Environmental Protection Agency 
Office of Pesticide Programs 
Registration Division (7505C) 

1200 Pennsylvania Avenue, NW 

Washington, DC 20460-0001 

Phone: (703) 305-5063 

E-mail: sweeney.kevin@epamail.epa.gov 


R. Vijayaraghavan, Ph.D. 

Defence Research and Development 
Establishment 
Jhansi Road 
Gwalior-474 002 
M.P., India 


Paul J. Weldon, Ph.D. 

Smithsonian Institution 
Conservation and Research Center 
1500 Remount Road 
Front Royal, VA 22630 
Phone: (410) 732-1539 
E-mail: weldonp@si.edu 

Graham B. White, Ph.D. 

University of Florida 

Entomology and Nematology Department 

P.O. Box 14565 

Gainesville, FL 32604-2565 

Phone: (352) 374-5968 

FAX: (352) 374-5922 

E-mail: gbwhite@ufl.edu 

Rui-de Xue, Ph.D. 

Anastasia Mosquito Control District 

500 Old Beach Road 

St. Augustine, FL 32080 

Phone: (904) 471-3107 

FAX: (904) 471-3189 

E-mail: xueamcd@bellsouth.net 


© 2006 by Taylor & Francis Group, LLC 



Table of Contents 


PART 1 Principles 

Chapter 1 History of Insect Repellents.3 

Sarah J. Moore and Mustapha Debboun 

Chapter 2 Terminology of Insect Repellents.31 

Graham B. White 

Chapter 3 Vertebrate Chemical Defense: Secreted and Topically 

Acquired Deterrents of Arthropods.47 

Paul J. Weldon and John F. Carroll 

Chapter 4 Human Emanations and Related Natural Compounds 

That Inhibit Mosquito Host-Finding Abilities.77 

Ulrich R. Bernier, Daniel L. Kline, and Kenneth H. Posey 

PART 2 Methods 

Chapter 5 Standard Methods for Testing 

Mosquito Repellents.103 

Donald R. Barnard, Ulrich R. Bernier, Rui-de Xue, and Mustapha Debboun 

Chapter 6 Biometrics and Behavior in Mosquito 

Repellent Assays.Ill 

Donald R. Barnard and Rui-de Xue 

Chapter 7 Animal Models for Research and Development of 

Insect Repellents for Human Use .125 

Louis C. Rutledge and Raj K. Gupta 

Chapter 8 Techniques for Evaluating Repellents.147 

John M. Govere and David N. Durrheim 


© 2006 by Taylor & Francis Group, LLC 













Chapter 9 Use of Olfactometers for Determining Attractants 

and Repellents.161 

Jerry F. Butler 

Chapter 10 Discovery and Design of New Arthropod/Insect 
Repellents by Computer-Aided Molecular 
Modeling.195 

Raj K. Gupta and Apurba K. Bhattacharjee 

Chapter 11 Molecular-Based Chemical Prospecting of Mosquito 

Attractants and Repellents.229 

Walter S. Leal 


PART 3 Products and Active Ingredients 

Chapter 12 Evaluation of Topical Insect Repellents and Factors 

That Affect Their Performance.245 

Scott P. Carroll 

Chapter 13 Repellents Used in Fabric: The Experience of the 


U.S. Military.261 

Wilfred C. McCain and Glenn J. Leach 

Chapter 14 Plant-Based Insect Repellents .275 

Sarah J. Moore, Annick Lenglet, and Nigel Hill 


Chapter 15 Considerations on the Use of Botanically-Derived 

Repellent Products.305 

Eugene J. Gerberg and Robert J. Novak 

Chapter 16 Efficacy and Safety of Repellents Containing Deet .... 311 

Stephen P. Frances 

Chapter 17 Lipodeet: An Improved Formulation for a Safe, 

Long-Lasting Repellent.327 

Buz Salafsky, Takeshi Shibuya, Yi-Xun He, Janies Ha, and 
Kalyanasundaram Ramaswamy 


© 2006 by Taylor & Francis Group, LLC 










Chapter 18 Picaridin.337 

Stephen P. Frances 

Chapter 19 DEPA: Efficacy, Safety, and Use of A,A-Diethyl 

Phenylacetamide, a Multi-Insect Repellent.341 

Shri Prakash, R. Vijayaraghavan, and K. Sekhar 

Chapter 20 PMD (p-Menthane-3,8-Diol) and Quwenling.347 

Daniel Strickman 

Chapter 21 IR3535 (Ethyl Butylacetylaminopropionate).353 

Germain Pnccetti 

Chapter 22 Older Synthetic Active Ingredients and 

Current Additives.361 

Daniel Strickman 

Chapter 23 Area Repellent Products.385 

Daniel Strickman 

PART 4 Uses 

Chapter 24 User Acceptability: Public Perceptions of 

Insect Repellents.397 

Stephen P. Frances and Mustapha Debboun 

Chapter 25 Commercially Available Insect Repellents and Criteria 

for Their Use.405 

Rui-de Xue, Arshad Ali, and Jonathan F. Day 

Chapter 26 Global Regulatory Perspective on Insect Repellent 

Development and Registration.417 

Kevin J. Sweeney 

Epilogue: Prospects for the Future .425 

Daniel Strickman, Stephen P. Frances, and Mustapha Debboun 

Appendices .429 


© 2006 by Taylor & Francis Group, LLC 













© 2006 by Taylor & Francis Group, LLC 




History of Insect Repellents 


Sarah J. Moore and Mustapha Debboun 


CONTENTS 

Historical Review.3 

Traditional Repellent Use Today.5 

Pyrethrum, Mosquito Coils, and Area Repellents.5 

The Development of Modern Synthetic Repellents.6 

Deet—A Breakthrough in Repellents.8 

Recent Repellent Discoveries.8 

DEPA.8 

IR 3535 .9 

Piperidine Compounds.9 

KBR 3023.9 

AI3-35765 and AI3-37220.10 

SS220.10 

Repellent Delivery Methods.11 

Area Repellents.13 

The Evolution of Repellent Testing.14 

Kairomones.14 

Choice.14 

In Vitro and Animal Tests.15 

Test Standardization.15 

References.17 


Historical Review 

It is likely that the use of repellents against biting arthropods developed thousands—possibly even 
millions—of years ago. Several species of primate have been observed anointing their pelage by rubbing 
it with millipedes and plants including Citrus spp., Piper marginatum, and Clematis dioica , 1-4 Wedge- 
capped capuchins ( Cebus olivaceus) were observed rubbing the millipede Orthoporus dorsovittatus 
onto their coat during the period of maximum mosquito activity. 5 The O. dorsovittatus species contains 
insect-repellent chemicals called benzoquinones, and it was hypothesized that the anointing behavior 
was designed to deter biting insects. Laboratory studies went on to show a significant repellent effect of 


3 


© 2006 by Taylor & Francis Group, LLC 



























4 


Insect Repellents: Principles, Methods, and Uses 


benzoquinones against Aedes ( Stegomyia ) aegypti (the yellow fever mosquito) 6 and Amblyomma 
americanum (the lone star tick). 7 Anointing behavior to deter blood-feeding arthropods is also 
common among birds, 8 and it may be genetically expressed as an “extended phenotype” because it 
has obvious adaptive benefit. 9 Evidence for this lies in the fact that benzoquinones applied to filter paper 
elicited anointing activity among captive-born capuchins. 6 

The first recorded use of repellents may be found among the writings of Herodotus (484 BCE—ca. 425 
BCE), who observed Egyptian fishermen. 10 Herodotus stated that, 

The Egyptians who live in the marsh-country use oil extracted from the castor-oil plant. This 
plant, which grows wild in Greece, they call Kiki, and the Egyptian variety is very prolific and has 
a disagreeable smell. Their practice is to sow it along the banks of rivers and lakes, and when the 
fruit is gathered it is either bruised and pressed, or else boiled down, and the liquid thus obtained 
is of an oily nature and quite as good as olive oil for burning in lamps, although the smell is 
unpleasant... 

It was argued that the oils acted as an area repellent because high densities of nuisance mosquitoes are 
active in the evenings in this region. This would have driven the Egyptians to their beds (where 
Herodotus also observed that they slept under rudimentary bed nets) had the lamp not provided 
protection from biting insects. 11 

The Romans also recorded methods of repelling flying insects (gnats) that would have included 
mosquitoes, as much of Italy was once swampland where the malaria vectors Anopheles labranchiae. 
Anopheles sacharovi, and Anopheles superpictus were abundant prior to the malaria eradication program 
of 1947. 12 The Geoponika is a collection of Roman agricultural lore, compiled during the tenth century 
for the Byzantine emperor Constantine VII Porphyrogenitus, that was heavily based upon the writings of 
Vindonius Anatolius (fourth century), as well as earlier writers, including Pliny. 13 The text suggests 
rubbing a concoction of vinegar, manna, and oil on the body, especially the head and feet, to repel 
gnats. 13 This may have had an effect on nuisance insects, especially mosquitoes, as natural vinegars 
contain acetic acid and smaller amounts of tartaric and citric acids. These acids may have had a mild 
antibacterial effect on the skin and therefore reduced the production of bacterial metabolites that 
mosquitoes use to locate human hosts, 14 particularly those produced by the feet. 15 In addition, some oils 
have a mild repellent action, 16 perhaps by reducing the emanation of host odor. In addition, Geoponika 
describes burning herbs such as black cumin ( Nigella sativa ), bay (Laurus nobilis), galbanum ( Ferula 
gummosa ), and oregano (Origanum vulgare ) to drive away nuisance insects. 13 Writings (ca. seventeenth 
century) derived from the ancient Sanskrit Yoga Ratnakara also contain references to the burning of 
plants to repel biting insects, including Vaca ( Acorns calamus), Marica ( Piper nigrum), asafoetidia 
(Ferula asafoetida), and Nimba or Neem (Azadirachta indica ). 17 Other remedies suggested in 
Geoponika and Yoga Ratnakara included burning fish, shells, various bones, dung, snakeskin, and 
peacock feather. This would have created a thick noxious smoke, as would have burning asafoetidia that 
has the colloquial name of Devil’s Dung in old French. This may have been perceived to work, as the 
smoke generated was thick and noxious to humans, although smoke does have some repellent effects on 
mosquitoes. 18 The smoke may mask human kairomones, particularly carbon dioxide, and the convection 
currents that mosquitoes need for short-range host location. Smoke production also lowers humidity by 
reducing the moisture-carrying capacity of the air. This makes mosquitoes susceptible to desiccation and 
reduces sensory input because mosquito chemoreceptors are more responsive in the presence of 

• 19 

moisture. 

In North America, native cultures relied heavily on plants, and many used plants to repel biting 
insects. 20 The Southern Carrier Tribe, or Dakelh, meaning “people who go around in boats,” live near 
rivers in British Columbia where mosquito densities are extremely high. This group used an infusion of 
common cow parsnip blossoms (Heracleum maximum) rubbed on the body to repel flies and mosquitoes; 
however, the more common mode of use was burning. For instance, the Colville Indians based around the 
Columbia River used leaves and stems of Common Yarrow (Achillea millefolium) as a smudge to keep 


© 2006 by Taylor & Francis Group, LLC 



History of Insect Repellents 


5 


away mosquitoes. The Blackfoot tribe, whose territory stretched along the Saskatchewan River, put the 
fringed sagewort ( Artemesia frigida) plant on campfire coals to drive away mosquitoes. Apparently, it 
was so effective that wild horses sheltered from insect pests in the smoke; consequently, the Indians used 
it to attract horses. The use of smoke against biting insects was carried on by European settlers as 
recorded by settlers to The Black Swamp in Ohio 21 : 

They [first settlers in Wood County] were subject to all kinds of deprivations. The most 
distressing of all the rest was their being subject to epidemics that swept through the country 
every summer and fall in the shape of malarial fevers... 

The warm months gave way to unrelenting swarms of gnats and mosquitoes... 

The most effective tool available [to fight the mosquito] was the smudge pot. These pots and their 
accompanying clouds of dark smoke discouraged the insects and were useful throughout most of 
the day; they were next to the cow while milking, under the table while eating, and even beside the 
bed while sleeping. 


Traditional Repellent Use Today 

Smoke is still the most widely used means of repelling mosquitoes utilized throughout the rural tropics. 
Waste plant materials are frequently burned in Sri Lanka as a mosquito repellent, even though indoor 
residual spraying has been carried out by the government for many years. 22 In rural Guinea-Bissau, 86% 
of residents used an unimpregnated bednet in conjunction with mosquito coils or plant-based smoke. 23 In 
the Solomon Islands, a recent survey revealed that fire with coconut husks and papaya leaves was the 
most prevalent form of personal protection from mosquitoes, being used by 52% of residents. 24 Surveys 
from South America found that 69 and 90% of respondents from Mexico 25 and Guatemala, 26 
respectively, burned waste materials to drive away mosquitoes. Smoke is also used to drive away 
biting insects in Southeast Asia: wood-fires and smudge pots are used in Myanmar, 27 whereas herbs are 
thrown on the fire in Yunnan, China. 28 

Although these methods are crude, many traditional repellents do have a repellent effect. A recent 
controlled field trial showed a comparable repellent effect produced by a 0.2% pyrethrin mosquito coil 
and lemon gum ( Corymbia citriodora) volatiles expelled by heating on metal plates. 29 Several field 
evaluations, where plants were burned to repel mosquitoes, have shown good reduction in mosquito 

• 'll 'll 

landings. One well-designed study in Papua New Guinea showed that burning local wood and 

leaves (mango wood, coconut husks, wild ginger leaves, and betelnut leaves) repelled between 57 
and 75% of mosquitoes. 31 Smoke also reduced indoor sand fly density by 1.7 times in East Africa. 32 

The use of smoke, although effective, requires continuous production in order to repel biting insects 
when used as an area repellent outdoors. 33 Although smoke does have a residual repellent effect when 
used within houses, 29 the indoor combustion of biomass has severe health consequences. 34 Therefore, 
safer and more modem methods of repelling mosquitoes are desirable. 


Pyrethrum, Mosquito Coils, and Area Repellents 

Pyrethrum is natural plant oil that occurs in the two species of pyrethrum daisy: Chrysanthemum 
cinerariifolium from the Dalmatian region and Chrysanthemum coccineum of Persian origin. The 
insecticidal component, comprising six esters (pyrethrins), is found in tiny oil-containing glands on 


2006 by Taylor & Francis Group, LLC 



6 


Insect Repellents: Principles , Methods, and Uses 


the surface of the seed case in the flower head. It is a highly effective insecticide that, although it has been 
used for centuries against all manner of insect pests, is relatively harmless to mammals. 35 

Pyrethrum is thought to have originally been used in China and was introduced to the Middle East 
along the trade routes through Central Asia, 36 from where it was introduced into Europe during the 
nineteenth century. 37 It is currently incorporated into mosquito coils to repel insects, and this practice 
probably derived from the incense used in religious ceremonies by Hindus, Buddhists, and the followers 
of Confucius. In Java today, the same incense used in ceremonies to honor ancestors is also used on a 
daily basis to repel mosquitoes. 38 

Pyrethrum powders were used by armies from the time of Napoleon to World War II to combat head 
and body lice. Before World War II, Japan was the major growing area, 37 and exported pyrethrum 
powder that was mainly used directly in its unrefined form as a powder for killing fleas. At that time in 
Japan, people usually mixed pyrethrum powder with sawdust and burned it in a brazier or incense burner 
to repel mosquitoes. Around 1890, the businessman Eiichiro Ueyama improved the pyrethrum powder 
and successfully developed a spiral-shaped mosquito repellent. 39 He formulated that idea when he met 
the son of an incense dealer at an inn in Tokyo. While talking with him, he came up with the idea of 
mixing starch powder with pyrethrum powder, then kneading it into the shape of stick incense. After 
several failures, Mr. Ueyama employed the workers of incense makers in Sakai, and thereby succeeded in 
creating a viable commercial product: a mixture of starch powder, dried mandarin orange skin powder, 
and pyrethrum powder. It was thoroughly mixed and kneaded, placed into a wooden mortar, extruded, 
and cut into the form of stick incense. Ueyama then replaced the wooden mortar with a compressing 
machine and was able to realize mass production. 

However, the bar-shaped mosquito stick burned rapidly, and several sticks had to be burned at once to 
obtain sufficient smoke to repel insects. In 1895, Yuki, the wife of Eiichiro, proposed making the stick 
thicker and longer, and curling it into a spiral shape. Eiichiro acted immediately on her proposal, but it 
was not until 1902, after years of experimentation, that he was finally able to complete a mosquito 
repellent with a spiral shape that was worthy of marketing. The final method involved cutting a thick bar 
of incense to a certain length and manually winding it. This same method continued to be used until 1957, 
when it was improved through machine punching, making mass production possible on a far larger 
scale. 39 Mosquito coils are widely used today: 29 billion mosquito coils are sold each year, 95% of them 
in Asia, 40 and household expenditure on these methods in the developing countries is substantial. 41 ' 42 
There is ample evidence that mosquito coils effectively repel mosquitoes. 43 

Pyrethrum affects the central nervous systems of all types of flying and crawling insects, blocking 
sodium-gated nerve junctions so that nervous impulses fail, 44 and the insect is knocked down and may 
die. In the lowest concentrations, pyrethrum affects insect behavior, producing a so-called “avoidance 
reaction” or “excito-repellency” that results in the insect fleeing the source of the chemicals. 45 
Synthetic analogues of pyrethrum were developed from the 1940s onwards. They exhibit a similar 
mode of action to pyrethrum, but are more potent and photostable. 46 ' 47 The insecticides broadly act in 
two ways: (1) the choreoathetosis/salivation (CS) pathway, and (2) the tremor (T) pathway. 48 
Importantly, these effects result in deterrency from entering a room where coils are burning, expellency 
of mosquitoes from within, interference with host-finding, bite inhibition, knockdown, and kill. 49 These 
repellence and bite-inhibition effects have been exploited to produce highly-efficacious repellents that 
combine permethrin (a synthetic pyrethroid) and deet, a synthetic repellent discussed extensively in this 
volume. 50 ' 51 


The Development of Modern Synthetic Repellents 

The military has conducted significant research into modern repellents to protect their troops from 
insect-borne disease. The first military repellents contained essential oils derived from plants. 


© 2006 by Taylor & Francis Group, LLC 



History of Insect Repellents 


7 



FIGURE 1.1 Repellents distributed to U.S. troops, Bombay 1945 (©Office of the Army Surgeon General, Public Affairs, 
and the Directorate of Information Management, Fort Detrick, MD, USA.). 


For instance, the Indian Army was issued a repellent comprised of citronella, camphor, and 
paraffin. 52 However, these repellents had limited duration, and intensive research began during 
World War II to find long-lasting repellents. The enormous burden of disease suffered by troops 
fighting in endemic areas motivated this research. For instance, 821,184 cases of malaria were 
recorded among U.S. troops involved in overseas campaigns, resulting in 302 deaths, 53 and over 12 
million lost duty days 54 With the advent of large-scale jungle warfare, chigger-borne scrub typhus 
became an important medical problem for troops in the Far East. Indeed, approximately 6,000 cases 
were to appear in U.S. forces alone during the campaigns that followed the outbreak of war with 
Japan. 55 Chiggers were also the cause of considerable discomfort for soldiers training in the U.S.; 
this resulted in the Surgeon General requesting the Orlando laboratory of the United States 
Department of Agriculture (USDA) to study means and methods for controlling chiggers by 
repellents or insecticides in the summer of 1941. 56 Between 1942 and 1945, over 7,000 potentially 
repellent compounds were tested by the USDA. 57 One of the first chemical repellents to be 
developed was dimethyl phthalate (DMP; it was patented in 1929 as a fly repellent), followed by 
Indalone® (butyl-3,3-dihydro-2,2-dimethyl-4-oxo-2H-pyran-6-carboxylate; patented in 1937), and 
ethyl hexanediol (2-ethyl-1,3-hexanediol), also called Rutgers 612, that became available in 
1939. 58 In 1942, DMP and Indalone demonstrated significant protection against chiggers when 
tested by Madden, Lindquist, and Knipling of the Orlando laboratory for troops in Louisiana. 59 The 
result was corroborated by field trials in New Guinea against scrub chiggers ( Leptotrombidium ) 60 
and a range of other species. 61 The introduction of chemical repellents dramatically lowered 
incidence of scrub typhus, 55 ' 60 and allowed less-restrictive battle uniforms. Prior to the introduction 
of DMP and Indalone for impregnation of uniforms and application to exposed skin, prevention of 
insect bites had relied on long clothing plus head nets and mosquito gloves. 56 The head nets were 
uncomfortably hot and restricted vision, making them unpopular with troops and therefore rarely 
used. The introduction of repellents for exposed parts of the body proved more popular 
(Figure l.l). 56 

After the war, a repellent known as 6-2-2 or M-250, containing 6 parts DMP, 2 parts Indalone, and 
2 parts ethyl hexanediol, became popular in the U.S.A. However, products containing ethyl hexanediol 
were voluntarily removed from the U.S. and Canadian markets in 1991 in response to an unpublished 
study by a manufacturer showing poor lung expansion in the offspring of exposed animals. 62 
Additional studies showed mild developmental toxicity after cutaneous administration to pregnant 
rats. 63 


© 2006 by Taylor & Francis Group, LLC 


Insect Repellents: Principles , Methods, and Uses 


Deet—A Breakthrough in Repellents 

Prior to the removal of 6-2-2 from the marketplace, its use was eclipsed by the discovery of deet 
(AW-diethyl-3-methlybenzamide or (V,(V-diethyl-M-toluamide) in 1953. 64 This was perhaps the single 
most important event in the evolution of repellents, and deet remains the principal, and the most effective 
repellent in use today 65 —more than 50 years after its discovery. Deet is a broad-spectrum repellent that is 
highly effective against all mosquitoes: Aedes spp., 66-69 including the dengue vectors Aedes aegypti 70-71 
and Aedes albopictus ’ ; Culex spp. ' - ; Mansonia spp. ’ ' ; and Anopheles malaria vectors, 

'J A 'JO "7Q OQ_QQ 

including the Afrotropical Anopheles gambiae, ’ ’ and Anopheles arabiensis, ~ Southeast Asian 
Anopheles dims, ' ' and Anopheles minimus ; South American Anopheles darlingi, and Western 
Pacific Anopheles farauti. 8687 Other insects of medical importance repelled by deet include sand flies 
(. Psychodiclae , both Old World and New World) - ; black flies ( Simulidae ) ’ ; chiggers (Trombicu- 
lidae) 92 ~ 94 \ hard and soft ticks ( Ixodidae) 95 ~ 98 \ bedbugs ( Cimex hemipterus) 99 ; and fleas 
(Siphonaptera). 100 It is, therefore, now used as the “gold-standard" repellent against which other 
substances are compared in laboratory and field trials. 

An estimated 15 million people in the United Kingdom, 78 million people in the United States of 
America., 1 1 and 200 million people globally use deet each year. 10- There has been much speculation 
on the safety of deet following reports linking it to seizures and encephalopathy, particularly in 
children, 103-106 as well as neurotoxicity, 107 especially in combination with other pesticides. 108 
However, deet has been used for 50 years with a tiny number of reported adverse effects, many of 
which had a history of excessive or inappropriate use of repellent. 109110 Nonetheless, its toxicology has 
been more closely scrutinized than any other repellent, but it has been deemed safe for human use, 101 ' 111 
including use on children 106 and pregnant women. 112 The use of a deet/permethrin repellent has recently 
been proven to reduce malaria incidence amongst users. 113 


Recent Repellent Discoveries 
DEPA 

Recently, DEPA (A,A-diethyl phenyl acetamide), a compound developed around the same time as deet, 64 
has received renewed attention. It has similar cosmetic properties to deet, similar dermal absorption and 
excretion, plus the symptoms of acute poisoning with DEPA are similar to deet. 114 However, its dermal 
toxicity to rats has been reported as LD 50 1.7-2.1 g/kg, 114 and 3-4g/kg, 115 which may require 
further clarification. 

In a field study, 0.3 mg/cm 2 DEPA in alcohol provided complete protection against Culex 
quinquefasciatus mosquitoes at a mean landing rate of 9.22 mosquitoes/person/h. 116 Another field test 
of DEPA with Culex quinquefasciatus, Simulium himalayense, and the leech Haemadipsa zeylanica 
showed 1.5, 2, and 1.5 h of complete protection, respectively. 117 However, control numbers were not 
given in this publication. Laboratory tests using rabbits showed that there was no significant difference in 
the response of the sand fly Phlebotomus papatasi to DEPA or deet. 90 Furthermore, in vitro application of 
repellents to a membrane blood feeding system, for Aedes aegypti, has shown that two analogues 
of DEPA, DM156 and DM34, show promising repellency and low toxicity, warranting further 
evaluation. 118 DEPA is an extremely cheap repellent, costing Rs. 1140 (U.S. $25.40) per kg compared 
to Rs. 2170 (U.S. $48.40) for deet. 116 This is because one of the precursors of deet, (3-methylbenzoic 
acid) is not readily available in India. 119 DEPA has now been formulated in a commercial preparation by 
the Defence Research and Development Establishment (DRDE) and has been granted approval by the 


© 2006 by Taylor & Francis Group, LLC 



History of Insect Repellents 


9 


Drug Controller of India. 120 This repellent may prove useful, particularly among residents of the 
developing world, for whom cost is the main motivator in personal repellent choice. 121 


IR 3535 

Insect repellent 3535 (IR 3535), P-fA-acetyl-V-butyl) aminopropionic acid ethyl ester], also known as 
MERCK 3535, was developed in 1975 by Merck, 122 and has been on the market in Europe for the past 
twenty years. It has low toxicity, although it is irritating to the eyes and sometimes the skin. 123 It became 
available in the U.S. in 1999 after being passed by the EPA, classified as a biopesticide, as it is a 
substituted B-amino acid, structurally similar to naturally occurring R-alanine. 124 

Efficacy data for IR 3535 is variable, but it is generally comparable with deet. Data from the laboratory 
showed IR 3535 to be equal to deet against Aedes aegypti, Culex quinquefasciatus, 69 ' 125 and Culex 
taeniorhynchus , but not Anopheles dints. 125 However, another laboratory study with Aedes aegypti and 
Anopheles maculatus showed IR 3535 to be significantly inferior to deet. 126 Field trials in Southeast Asia 
against Aedes albopictus, and Culex gelidus 125 ; and Aedes albopictus, Culex quinquefasciatus, and Culex 
bitaeniorhynchus 127 found that IR 3535 and deet offered similar protection. However, a test against 
Aedes cantons and Aedes annulipes under initial biting pressures of 714 landings/person/h produced data 
that indicated that deet had a duration twice that of IR 3535 (4.8 vs 9.7 h). 128 A further test against Aedes 
(Ochlerotatus ) taeniorhynchus in the Everglades, also under high biting pressure, measured no 
significant difference between the protection offered by deet and IR 3535. 68 A comprehensive field 
test against Anopheles gambiae showed that IR 3535 decayed at a similar rate to deet, 78 and the World 
Health Organization (WHO) has declared it a safe and effective repellent for human use. 123 In fact, there 
is not a single recorded incidence of an adverse reaction to this compound. 

Piperidine Compounds 

There has been a flurry of renewed interest in the piperidine-based compounds, leading to the discovery 
of several new and highly effective repellents. Piperidines, as a chemical class, are cyclic amines. The 
piperidine skeleton is present in piperine, the main active chemical agent in pepper ( Piper sp.). During 
the 1970s, approximately 600 synthetic compounds related to piperidines were developed by scientists at 
the Gainesville and Beltsville research centers of the USD A. The data from these experiments is now 
being re-examined using new, high-tech methodologies coupled with rapid-screening bioassays. This 
interest in finding deet alternatives has been motivated by the controversy around the safety of deet, its 
low user acceptability, and its plasticizing effect. 


KBR 3023 

The repellent 1-piperidine carboxylic acid-2(2-hydroxyethyl)-l-methylpropylester was developed by 
Bayer in the 1980s using molecular modelling. 129 It has several synonyms: Picaridin is its common 
name, Bayrcpcf' is its Bayer trademark name, Icaridin was used by WHO, and KBR 3023 is another 
trade name. This compound, the most recent piperidine derivative, is registered in many European, South 
American, Asian and African countries as well as Japan, Canada, and the U.S. Its most important new 
feature is its very low toxicity (EPA Grade III). Most importantly, it elicits practically no dermal or eye 
irritation (EPA Grade IV) nor skin sensitization. 130 Furthermore, it does not have a significant 
plasticizing effect, which is a major drawback of deet. Cosmetically, it is superior to deet as it is 
colorless, odorless and has a pleasant feel on the skin. 131 A user acceptability study showed a distinct 
preference for KBR 3023 among Australian troops when compared to deet, which was uncomfortably 
oily or caused irritation to half of respondents. 132 

The efficacy of Picaridin is excellent, and it is generally superior to deet in terms of longevity. In a 
carefully designed field evaluation against Anopheles gambiae and Anopheles funestus, KBR 3023 in 


2006 by Taylor & Francis Group, LLC 



10 


Insect Repellents: Principles , Methods, and Uses 


ethanol outperformed deet after a 10-h exposure, and the half-life of the repellent was one hour longer 
than that calculated for deet and IR 3535, 78 using modelling first used by Rutledge et al. in 1985. 133 This 
is because KBR 3023 evaporates at a slower rate than deet. Were it not for the lower volatility of 
Picaridin, it would probably be less effective, because dose for dose it is less repellent than deet when 
freshly applied. 134 Other studies have shown a similar performance when compared to deet in field trials: 
against Anopheles spp. 76 and Verrallina lineata 87 in Australia, Aedes ( Ochlerotatus ) taeniorhynchus in 
U.S., 8 Aedes albopictus , Culex quinquefasciatus and Anopheles spp. in Malaysia, 1 3 as well as one field 
trial under biting pressures of 1,200-2,400 Aedes cantans and Aedes annulipes landings/person/h. 128 
KBR 3023 has also shown similar efficacy to deet against Aedes aegypti, Anopheles gambiae , 136 and 
Amblyomma hebraeum 98 in laboratory tests. 

It is this combination of efficacy, safety, and cosmetic appeal that has led to the WHO designating 
KBR 3023 as its “repellent of choice for malaria prevention.” 137 In addition, the Centers for Disease 
Control and Prevention (CDC) recommended both deet and KBR 3023 for West Nile virus and malaria 
prevention. 138 It is also being investigated for incorporation into military repellents after outperforming 
the standard Australian Defence Force formulation of 33% deet. 87 ' 132 


AI3-35765 and AI3-37220 

The piperidine compounds l-[3-cyclohexen-l-ylcarbonyl] piperidine, called AI3-35765, and 
l-[3-cyclohexen-l-ylcarbonyl]-2-methylpiperidine, also known as AI3-37220, were first synthesized 
by the USDA in 1978. 139 It should be noted, however, that neither of these compounds is 
available commercially. 

Research on AI3-35765 showed it to have similar efficacy as deet against Anopheles albimanus. 
Anopheles freeborni. Anopheles gambiae, Anopheles stephensi, and Phlebotomus papatasi 88 ', Prosimu- 
lium mixtum, and Prosimuliumfuscum 1 ; Anopheles stephensi and Culex quinquefasciatus 14 ; as well as 
Culex pipiens, both in the laboratory and the field. 88 A13-35765 was dropped from the Army research 
program, despite its impressive efficacy, because it caused an uncomfortable liniment-like warming 
reaction on some peoples’ skin (Dan Strickman, pers. com.). However, recent interest has focused on 
AI3-37220, a compound consisting of a racemic mixture of four isomers. 142 This mix has proven highly 
effective against a variety of blood-feeding arthropods, including Anopheles albimanus, Anopheles 
freeborni. Anopheles gambiae, Anopheles stephensi , and Phlebotomus papatasi 88 ', Prosimulium mixtum 
and Prosimuliumfuscum 1 : and Aedes communis and Simulium venustum. In fact, its longevity was 
shown to be superior to that of an equivalent concentration of deet in field trials with Anopheles farauti in 
Australia 143 and Papua New Guinea, 86 Anopheles dims, 144 Anopheles funestus and Anopheles 
arabiensis , 80 Leptoconops americanus , 145 Amblyomma americanum 96 and a laboratory trial with 
Anopheles stephensi. It has undergone extensive toxicology testing and has been deemed safe. 1 ' 7 
However, it should be noted that it has not yet undergone all of the necessary toxicological testing to 
support registration. 


SS220 

The latest development in synthetic skin repellents is optically active (lS,2S)-2-methylpiperidinyl-3- 
cyclohexen-1-carboxamide, discovered by the USDA and dubbed SS220. It is derived from AD- 
37220—insomuch as it is the most repellent of the four isomers that comprise racemic AI3-37220, and is 
2.5 times as effective as the racemic mixture against Aedes aegypti. 148 Laboratory tests showed SS220 to 
be equal to deet against Anopheles stephensi and Aedes aegypti, and better than KBR 3023 against Aedes 
aegypti. 144 Against the tick, Ixodes scapularis, SS220 outperformed deet and was as good as deet against 
Amblyomma americanum. 149 However, SS220 is less effective than deet against Anopheles albimanus. 150 
To date, no field studies have been published, although a USDA report stated that SS220 equals the 
effectiveness of 33% deet. 151 It has been reviewed by the U.S. military as the new active ingredient to 


© 2006 by Taylor & Francis Group, LLC 



History of Insect Repellents 


11 


replace deet. Extensive toxicological tests have shown low irritation and toxicity. 152-156 In addition, 
SS220 has a low rate of evaporation that will improve longevity. Using a smaller amount of long-lasting 
repellent makes for a more cost-effective and safe product because potential dermal absorption will be 
reduced. User acceptability is also likely to be higher because it has a slightly fruity odor, does not have 
an oily consistency, and has little plasticizing effect. 157 The disadvantage of SS220 lies in the fact that it 
is a single stereoisomer, and will, therefore, be more costly to produce than a racemic mixture. 
Furthermore, SS220 has not yet been registered, and the huge costs associated with this process, 
although necessary, mean that many promising new compounds may never be realized, as developers 
need to consider the potential financial benefits of registering a compound versus the initial large 
monetary outlay. 


Repellent Delivery Methods 

Most insect repellents are effective in the vapour phase, defined as vapour or olfactory repellents by 
Garson and Winnike as “those materials which are sufficiently volatile to keep an insect at a distance.” 158 
As the repellent molecules are volatile, temperature, humidity and wind affect evaporation of the 
repellent and therefore its longevity. 159-161 Perspiration and abrasion will also reduce the longevity of the 
repellent. 162 ' 163 

Many effective repellents have a high vapor pressure and are therefore volatile. At high mosquito 
densities, a heavy dose of a low vapor pressure repellent may be necessary to repel mosquitoes initially, 
whereas repellents with high vapor pressures may offer protection at low concentration. Subsequently, 
the lower evaporation rate of a repellent with less volatility means that it will continue to repel for a 
longer time period. 164 For instance, citronella (Cymbopogon nardus) essential oil and pure deet have 
similar ED 90 values of 112.8 and 95.5 nE/cm 2 , respectively, for Anopheles gambiaedes. 14 However, pure 
citronella oil at a dose of 3.33 X 10 -3 mL/cnr provides protection for only 2 h. 165 Citronella contains 
actives such as citronellol that has a vapor pressure of 0.009 kPa vs. 0.003 kPa at 20°C for deet. 166,167 

The high rate of loss of repellents was overcome by using extremely high concentrations of deet, 
especially among military personnel. Standard deet concentrations for military repellents were 75% 
(U.S.) and 95% (Australia). 132 However, repellents may also be lost through dermal absorption. 
Absorption of deet is generally high, at 0.8%/h in humans. 168 The high rate of dermal absorption 
raised safety concerns for adverse side effects associated with using high concentrations of deet. This 
prompted several collaborative research studies that eventually resulted in the development of slow- 
release formulations based on creams, polymer mixtures, or microcapsules that are available on the 
market today. Increased repellent longevity may be achieved in one of three ways: (1) controlling release 
or lowering vapor pressure, (2) preventing or reducing repellent absorption, or (3) improving resistance 
to abrasion and sweat. 

Formulations can prolong the effect of repellents. Initially, additives such as olive oil 169 and mineral 
oil 170 were used. They may improve repellent longevity by inhibiting loss of repellent volatiles and loss 
through sweating and abrasion. 17 Additives such as perfume fixatives were also researched. Fixatives 
are large branching molecules that lower the vapor pressure of repellents. These included Tibetene and 
vanillin, both of which have a significant effect on repellent longevity, increasing it by 29 and 95%, 
respectively, when used with deet at a 1:1 ratio. 172173 

In the early 1970s, an intense research program involving military, federal, academic, institutional, 
and industrial investigators began with the aim of providing a non-toxic, cosmetically-acceptable, and 
effective repellent system that would repel insects for 12 h under tropical conditions. They aimed to 
develop a repellent that would provide 24-hour protection under conditions that induced sweating 
through a two-pronged approach: (1) searching for agents with higher intrinsic repellency and 


2006 by Taylor & Francis Group, LLC 



12 


Insect Repellents: Principles, Methods, and Uses 


(2) enhancing repellent protection time of deet using hydrophobic agents to maintain the repellent on 

i • 174 

sweating skin. 

The research on development of a “binding agent” was carried out by the Letterman Army Institute of 
Research (LAIR). The formulation of deet with film-forming polymer resins was aided by enlisting the 
help of industrial and institutional laboratories. Evaluation of cosmetic properties, dry-skin protection 
time, and wash resistance was carried out at LAIR using radio-labelled deet formulated with polymer- 
film formers to study evaporation, skin penetration, and wash resistance. 174 During the first year of skin 
testing, several formulations were developed that were far superior to deet in both dry protection time and 
wash resistance, but few were cosmetically acceptable. However, the basic premise that film formers are 
extremely effective in enhancing protection time was confirmed. Then, almost a year was spent 
attempting to upgrade the cosmetic properties of those formulations that had superior wash resistance. 
The majority of research used silicone and carboset acrylic polymers and showed dramatically enhanced 
protection times. One example was the use of silicone that improved the dry protection time of deet by a 
factor of two, although it did not impart appreciable wash resistance. Tests with carboset acrylic 
polymers enhanced the dry protection time of deet and significantly improved wash resistance. Over 150 
reformulations were prepared, examined, and about half were studied for wash resistance and cosmetic 
appeal on volunteers. However, little success was realized: cosmetically-elegant formulations had 
inferior wash resistance, whereas systems having superior wash resistance were sticky or brittle on the 
skin. A further year was spent expended in attempting to reformulate carboset polymers to improve their 
cosmetic appeal without sacrificing their excellent wash resistance. Formulations of carboset/deet were 
combined in increments with silicone polymer (decreasing carboset content in each member of the 
series) trying to upgrade cosmetic acceptability without losing the excellent wash resistance. 174 

This research at LAIR in the late 1970s and early 1980s established the physical parameters and 
theoretical framework that demonstrated the feasibility of polymer and microcapsule mechanisms to 
release deet at a predetermined rate. The formulations tested in those early studies utilized microcapsule 
and polymer systems designed to provide continuous long-term release of the active ingredient. In 
microcapsule formulations, the active ingredient is contained in tiny capsules produced by coacervation, 
interfacial polymerization, extrusion, and other processes. The release rate is determined by the size and 
number of the microcapsules, the composition and thickness of the microcapsule walls, the concentration 
and properties of the excipient, and other additives used. These formulations may also contain free active 
repellent in addition to that contained in the microcapsules. In polymer systems, the active ingredient is 
formulated with a polymer that will form a thin film over the skin. This film acts as a reservoir for the 
active ingredient and slows its absorption and evaporation. In microparticulate controlled-release 
systems, the active ingredient is absorbed on the surface of microparticles and released slowly over 
time. 70 Further research was conducted that looked at formulations based on hydrophilic vinyl polymer, 
polyvinylpyrrolidone (PVP), 133 before the 3M Corporation’s proprietary polymer formulation was 
finally devised. The polymers and microcapsules in the formulations slow the absorption and evaporation 
of deet, thereby holding it on the surface of the skin, where it can continue to repel arthropods for an 
extended period of time. 171 

Ultrathon™ (3M) has been the military topical repellent of choice since 1990, when it first became 
available in the military supply system. The product contains 33% deet in a controlled-release polymer 
base, and is a nongreasy, white lotion with a mild, pleasant odor. 175 It was validated by the USDA 176 and 
chosen as a result of tests against a variety of mosquito species under three climatic regimes: (1) 24°C 
and 98% relative humidity (RH), (2) 30°C and 78% RH, (3) 37°C and 31% RH. 70 In these tests, the 
polymer formulation performed as well as a microparticulate formulation of 42% deet (Biotek) or 75% 
deet in alcohol (former standard of the U.S. Army). Field trials in the Philippines with Anopheles 
flavirostris showed that 3M was significantly more effective than 71% deet in ethanol for between 6 and 
12 h after application. 177 However, in Australian field tests against Anopheles farauti, 3M and Biotek 


A registered trademark of 3M Corporation, Minneapolis, MN. 


© 2006 by Taylor & Francis Group, LLC 



History of Insect Repellents 


13 


performed as well as the Army deet formulation. 176 In this study, the volunteers applied repellents 
themselves according to label instructions, therefore reflecting normal use conditions. There was a 
significant difference in the amount of product applied by the individuals, but due to differences in the 
deet concentration of the three formulations, the amount of deet applied was fairly consistent. 176 The 3M 
33% deet polymer formulation was found to be just as effective in repelling mosquitoes in field tests by 
the Australian military, 87 ' 178 and a 35% deet formulation in cellulose gel that lowered dermal absorption 
and evaporation of deet was placed into service in Australia in 1992. 132 The British military now also 
uses 3M Ultrathon. 179 Importantly, the slow-release formulations have significantly lowered dermal 
absorption, compared to ethanol formulations with deet at comparable concentrations. 180 The addition of 
polymers also improves the cosmetic appeal of repellents by lowering the amount of deet available, thus 
reducing odor, stickiness, and plasticization, as well as improving abrasion and wash resistance. 181 
Several polymer and microencapsulated formulations are available on the market, including Sawyer 
Controlled Release®, HourGuard®, Skedaddle®, and Ultrathon. 11()J 

Gel-based slow-release deet formulations, such as Ultrathon™, have low acceptability among troops. 
Recently, 10% of American soldiers serving in Kuwait, Haiti, and Bosnia used the U.S. Army repellent 
containing 33% deet alone, 29% used commercial formulations, 34% used both types, and 27% used 
neither. 182 A similar situation was witnessed among Australian troops. Only 26 out of 955 soldiers 
interviewed used the standard issue 35% deet formulation in gel base. 183 The main reason given for 
nonuse is the sticky feel of the repellent on the skin. 183 Soldiers do not use military-issued repellents for 
several reasons, including their previous familiarity with nonmilitary products before joining the 
military, availability of commercial options, and advertising of repellents in various commercial 
media. 182 Additionally, soldiers’ perceptions of what is acceptable or good has been demonstrated by 
Garnbel et al. (1998), 182 who observed American soldiers declining free military issue (33% deet 
formulation) repellent in an olive-green tube for a commercial product that was identical to the military 
issue, except that it was packaged using a different name and supplied in a brightly colored tube. 

Area Repellents 

There has been a recent increase in interest in area repellents that repel all biting insects within a set 
distance of the source of repellent molecules. Mosquito coils that are area repellents continue to be the 
most popular form of personal protection in use today. 40 In addition, citronella candles are commonly 
used as insect repellents in backyards and can provide 42% protection. 184 Spatial repellents have been 
defined as “an inhibiting compound, dispensed into the atmosphere of a three dimensional space which 
inhibits the ability of mosquitoes to locate and track a target such as a human or livestock.” 185 As 
repellents act in the vapor phase, they may potentially have a long-range effect through toxicity or 
confusing signals that indicate the presence of a host, established by saturating a zone or space with the 
spatial repellent. 186 One important concern with area repellents is the fact that they may only be used 
under conditions where air flow is minimal—for instance, in forests—as the repellent volatiles may be 
diluted with significant air flow. 28 

A new development in spatial repellent technology is the ThermaCELL® * 1 Mosquito Repellent system, 
consisting of a butane-fueled generator that heats a metal plate to volatilize cis or trans-allethrin from an 
impregnated pad. The repellent is effective over a distance of 7 m and provided > 90% protection over 
6 h from sand flies and mosquitoes in a field trial in Turkey. 187 The system vaporizes the active ingredient 
from paper mats that are heated. These are highly effective, even under drafty conditions, as shown by 
laboratory and field trials. The various devices that heat such impregnated mats are second only to 
mosquito coils in global consumption. 40 The ThermaCELL system is an excellent development in 


Sawyer Controlled Release is a registered trademark of Spectrum Brands; HourGuard is a registered trademark of 3M 
Corporation; and Skedaddle is a registered trademark of Multicrop International Pty. Ltd. 

1 A registered trademark of Schawbel Corportation, MA. 


© 2006 by Taylor & Francis Group, LLC 



14 


Insect Repellents: Principles , Methods, and Uses 


repellent research, as it allows several people to be protected at once. However, it costs approximately 
U.S. $1.00 per hour to run, making it too expensive for use in the developing world. 

Several experiments with y-port olfactometers have recently shown the repellent effect of plant-based 
components such as the essential oil of catnip. 190 Plant-based repellents usually have a short longevity 
when applied to the skin, as they have high vapor pressure. 166 ' 191 However, it is this feature that makes 
them excellent spatial repellents. In a field test, mint ( Mentha arvensis) oil volatilized using a kerosene 
lamp significantly protected volunteers from Mansonia titillans , 28 and field experiments have demon¬ 
strated the spatial repellent effect of volatilization of plant oils using heated plates against Anopheles 
gambiae. 29 


The Evolution of Repellent Testing 
Kairomones 

Progress in the development of new repellents has been slow until the recent breakthroughs, perhaps due 
to improved understanding of the repellents’ modes of action on the target organisms. However, this is 
now changing, and many papers have now been published on the mode of action of host kairomones on 
host-seeking insects. 14 - 192 ~ 202 Delicate methods, such as electroantennogram readings of the response of 
sensory neurons in insect antennae to attractive and repellent compounds, have allowed greater 
understanding of insects’ sensory systems. ’ “ Y-port olfactometers have also proved very useful 

in discriminating the effect of kairomones, as well as insect repellents and inhibitors. They have 

shown the importance of the interplay between whole host odor and repellents. Olfactometer experiments 
with deet have shown that it is not a true repellent insomuch as it causes insects to make oriented 
movements away from its source, but it is an inhibitor that prevents insects from feeding on a host in its 
presence (in this case a host-derived odor blend). 186 ' 214 Many groups are working on quantifying what 
elements of human skin and breath are actually attractive to host-seeking insects, and, in particular, 
highly-anthropophilic disease vectors. Cork and Park (1996) 215 chemically fractionated human sweat 
samples into acid and nonacid components. They measured the electrical response of sensillae in the 
antennae of mosquitoes, and found that short-chained aliphatic acids (C 2 -C 8 ) elicited significantly 
greater responses than the longer-chained acids. These acids elicit a landing response 198 and they have a 
significant effect on mosquito host-seeking behavior. 203 There is a growing body of evidence indicating 
short-chained fatty acids are reliable cues; however, these require complex blends, including synergists 
such as ammonia and lactic acid. 206-216 Therefore, the potential for an olfactometer with a reliable 
synthetic lure for repellent testing is some way off. 

Choice 

Should olfactometers become used regularly in the future for repellent/inhibitor screening, they will only 
ever be suitable for preliminary screening because olfactometers allow insects to choose between one or 
more targets. This is a disadvantage, as it causes an inflation of repellent efficacy: it “shifts the point of 
reference for the ED 50 to a lower level.” 217 It was argued that “free choice” between repellent-treated and 
untreated areas more accurately reflects use conditions where mosquitoes will feed on untreated areas of 
a repellent user, or their untreated companions. 218 This, however, is not a useful scenario, as a single 
infected bite is sufficient to transmit vector-borne pathogens. Therefore, recent publications have stressed 
the importance of high (>95%) protection, where the mosquito has no choice but to feed on repellent- 
treated skin if they wish to feed at all. 78 ' 219 In addition, several experiments have demonstrated that 
offering mosquitoes free choice in laboratory overestimates repellency. 74 ' 220 A free-choice test 
calculated the ED 50 of deet as 0.024-0.042 mg/cm 2 , 221 whereas a similar test with no choice calculated 
it as 0.35 mg/cnr. “ In field tests, Barnard et al. showed that the application of a repellent to one limb 


© 2006 by Taylor & Francis Group, LLC 



History of Insect Repellents 


15 


where the other was used as a control inflates repellency, while the use of repellent using a repellent- 
wearing “bait” individual with an untreated “collector” also overestimates repellency. 222 A study in 
Vietnam was performed with Anopheles dims using one pair (one wearing deet, and one solvent control) 
and one individual (wearing deet) sitting alone. 223 In this case, the repellent wearer that was sitting alone 
received 3.5 times more mosquito bites than the repellent wearer that was sitting close to an alternate 
blood source. This is because the mosquitoes will always feed on the “easiest option” with least repellent, 
be that an adjacent area of skin, an alternate limb, or another individual. When the protection afforded by 
the repellent wanes, mosquitoes will start to feed through the repellent. However, if there is an 
unprotected alternative, they will be diverted and feed upon it. This also applies in field tests if 
individuals are less than 10 m apart because this is the limit of short-range attraction. 224 

In Vitro and Animal Tests 

Tests on repellents, from the 1920s until recently, were often performed on shaven animals including 
rabbits, dogs, guinea pigs,“ and chicks. This method, despite questions regarding the ethical 
treatment of animals, may distort the results of repellent tests. Nicolaides et al. (1968) 228 compared the 
skin of humans and other domestic animals. They concluded that humans excrete mainly triglycerides and 
are, therefore, unique in having fatty acids as breakdown products on the skin surface. This means that 
short-chained aliphatic acids are reliable host cues for anthropophilic mosquitoes, and, therefore, testing 
repellents on animals will not give representative data of how the repellent will perform when applied to 
human skin. In addition, the most efficient malaria vectors are extremely anthropophilic, and will be less 
attracted to nonhuman hosts, possibly due to genetically mediated innate preferences. Thus, this 
method gave a distorted measure of repellency. Indeed, Rutledge et al. directly compared measurements of 
repellent efficacy obtained using rabbits 171 and mice 230 with that obtained using human arms. In both 
cases, repellents showed greater variability and greater persistence when applied to animals than humans. 

Other studies have utilized membrane blood feeders, commonly used for feeding mosquitoes in 
insectaries, to measure repellency. 217 Although the data obtained using this method roughly correspond 
to data obtained with human-arm tests, 220 this method should be used only for rapidly screening large 
numbers of repellents to narrow down candidates for further testing. This is because membrane feeder 
tests differ from human-arm tests because mosquitoes do not respond as enthusiastically to a feeder as 
they do to a living host, and there is much interspecific variation in readiness to feed from membrane 
feeders. 231 Another testing method employs disks of paper impregnated with a test repellent, and the 
numbers of insect landings on impregnated and unimpregnated control disks are counted. This was 
shown to be an excellent method for testing irritancy of a chemical, but is not a measure of repellency. 232 
In vitro methods are cheap, and yield many results rapidly with no risk to human subjects, but they do not 
accurately mimic the conditions of repellent usage. Thus, different methodologies cannot be compared, 
nor can their results be directly extrapolated to the end user. This is particularly important since the 
discovery that deet, the leading insect repellent, is an inhibitor and not a true repellent. 214 

Test Standardization 

Other recent developments in repellents research have followed after the call of the WHO in 2000 233 to 
standardize repellent testing protocols. It would appear that tests are becoming far more stringent 
and standardized. 

It is always preferable to conduct tests on human volunteers for greatest accuracy, provided that 
laboratory-reared mosquitoes are used to eliminate the risk of pathogen transmission, and the selected 
volunteers show mild or no allergic reaction to mosquito bites. 234 It is conventional to use Aedes aegypti 
mosquitoes for repellent testing, but people generally show milder reactions to Anopheles bites. Aedes 
aegypti are commonly used, as they are easy to rear under laboratory conditions, and are avid biters. 
However, several other species also fulfill these criteria, including Anopheles stephensi. Anopheles 


2006 by Taylor & Francis Group, LLC 



16 


Insect Repellents: Principles , Methods, and Uses 


gambiae s.s.. Anopheles arabiensis , and Anopheles albitarsis. The U.S. EPA now recommends using 
Aedes aegypti along with a representative human biting species from both the Anopheles and Culex 
genera for laboratory studies of repellent efficacy. 235 It is preferred to perform bioassays on the vectors in 
the region for which the repellent is to be used 233 because the sensitivity of different mosquito species to 
repellents varies. 74 In addition, as deet is the active ingredient of most commercially available skin 
repellents and is the most effective and well-researched insect repellent available at this time, it is now 
considered a useful standard against which the effectiveness of alternative repellents may be judged. 234 

Laboratory tests are generally conducted with mosquitoes held in large laboratory cages into which the 
forearm(s) of the volunteer is introduced with the hand protected by a glove. The whole forearm may be 
exposed, or a 25 cm 2 area of skin, the remainder being covered with a rubber sleeve. In some tests, the 
repellent is applied to a cotton stocking, as repellents are much more persistent on fabric than on skin 
because loss of repellent through abrasion, skin absorption, evaporation and sweating is reduced. The 
stocking is drawn over another stocking that has been drawn over the arm to prevent skin contact with a 
repellent compound; this is particularly important for volunteers involved in regular testing, or where 
compounds have not been screened for toxicity or dermal absorption. However, it was shown that this 
method does not correlate well with results from tests where repellent is applied directly to the skin, 236 
and further studies need to be performed when the substance is deemed safe for use on the skin after 
toxicological evaluation. 

A major source of variation in laboratory tests is caused by differences in mosquito avidity related to 
their physiological state. The team from the USDA Agricultural Research Service’s Mosquito and Fly 
Research Unit have made excellent progress toward standardization of repellent testing methodology. 
They have published a series of important papers showing that mosquito attack rates, and consequently 
repellent protection times, are significantly influenced by mosquito body size (hence larval nutrition), the 
age and parity of the mosquitoes, as well as the time of day." “ Allowing the mosquitoes access to 
sugar solution or blood will also decrease their avidity, and subsequently, the measured repellency of a 
chemical because they will be at least partially engorged. 241 Also of importance is the density of the 
mosquitoes in the cage. For "time to first bite,” it was shown that the most rigorous tests required 
densities of mosquitoes where each mosquito had 49 cm 3 . 239 These experiments have drawn the 
conclusion that mosquitoes used for repellent testing should therefore be nulliparous, aged between 3 
and 10 d, and denied access to sugar prior to testing repellents. 

In addition, the EPA’s FIFRA (Federal Insecticide and Rodenticide Act) Scientific Advisory Panel has 
advised that the commonly used “time to first bite” test should no longer be utilized. 219 This test was 
commonly used where one arm is treated with 1 mL of a 25% solution of the test compound in ethanol. 
The arm is exposed for 3 min in every 30 min and the first time after treatment noted at which a bite 
occurs followed by a “confirmatory” bite in the same or the following exposure period. However, it has 
been concluded that the time to first bite method has not been developed using a statistically valid 
approach because its result depends on the behavior of a few individuals in the upper distribution of 
tolerance and does not reflect the behavior of the population as a whole. Therefore, it is increasingly 
recognized that methods that measure 95% reduction in bites are preferable because all bites are counted 
and the method provides a more “real-world” assessment of insect repellent efficacy. 219 This method 
requires sequential exposure of an arm with zero, and then progressively higher, doses of repellent for 
30 s to cages containing approximately 50 hungry Anopheles gambiae (or 45 s with Anopheles 
Stephensi). The number biting at the end of the short exposure is quickly counted (preferably with the 
help of an assistant) and the mosquitoes are then shaken off before they can imbibe any blood. Hence, 
the same mosquitoes can be used for testing each dose, and their continued hunger can be checked by 
exposing the other untreated arm. Probit analysis is used to calculate the ED 50 , ED 90 , or ED 95 . After 
reaching a dose that gives 100% repellency, the arm is re-exposed hourly until repellency declines to 
50% compared with contemporary counts on the untreated arm to measure the duration of this protection. 
This is a labor-intensive method of repellent evaluation, but it allows the direct comparison of repellents 
via the ED values. It also measures the relative tolerance of different species to repellents. 


© 2006 by Taylor & Francis Group, LLC 



History of Insect Repellents 


17 


A new, less labor-intensive method that may be used to calculate the probit curve of a repellent is the 
K&D module. 242 The K&D module is made of plexiglas and has six cells. Each cell has a stoppered 
access hole for transfer of mosquitoes to the cell, and a bottom with a rectangular 3 cm X by 4 cm hole 
that is opened and closed by a sliding door. 

The concave bottom conforms to the curvature of a human thigh and a separate bottom section with the 
same dimensions serves as a skin-marking template. A human test subject wearing shorts, seated with 
legs horizontally extended, uses the template and a water-soluble marker to denote the areas to which 
repellent or control is applied that correspond to the openings of the module. The test compounds are 
exposed to mosquitoes by placing the K&D module over it and opening the cell doors. The number of 
insects biting in each cell within the 2-rnin exposure is recorded, after which the doors are closed. At the 
conclusion of each assay, mosquitoes are freed by opening cells of the K&D module in a sleeved, 
screened cage. The method can be used to calculate the ED 90 of a compound by applying incremental 
doses of repellent, and may prove to be a simple and efficient rapid screening tool. 

Although laboratory tests are extremely useful, it is now generally agreed that field tests are the 
definitive test of a repellent, as they allow the evaluation of a substance under representative user 
conditions.” ’ ' Field tests allow the evaluation of a repellent with the desired test species, under 

the environmental conditions that it will be required to perform. Tests that use a Latin square or “round 
robin” design are favored to ensure that an adequate number of replicates are employed, as individual 
attractiveness to hematophagous insects, as well as their ability to capture them in tests, varies 
widely. 219 

Repellent science has advanced greatly in the last decade and will continue to progress in the future. 
New methods such as the molecular modelling and characterization of repellent molecules that attempt to 
explain the structure-activity relationship of repellent molecules, especially their stereochemical activity 
relationships, are beginning to emerge. 243 In the future, new repellents may be discovered based on their 
molecular structure and tested in the field using host-odor-baited traps. This will remove the risk 
associated with testing repellents against the insect vectors of pathogens, the infection with which they 
are designed to prevent. Until then, the recent advances in laboratory science mean that the rapid and 
accurate screening of candidate repellent compounds that closely represent field conditions is becoming 
more and more attainable. 

References 

1. M. Baker. Fur rubbing: Use of medicinal plants by capuchin monkeys ( Cebus capucinus), American 
Journal of Primatology, 38, 263, 1996. 

2. C. R. Birkinshaw, Use of millipedes by black lemurs to anoint their bodies, Folia Primatologica, 70, 
170, 1999. 

3. D. J. Overdorff. Similarities, differences, and seasonal patterns in the diets of Eulemurfulvus rufus and 
Eulemur rubiventer in the Ranomafana National Park, Madagascar, International Journal of 
Primatology, 14, 721. 1993. 

4. M. Zito. S. Evans, and P. J. Weldon, Owl monkeys ( Aotus spp.) self anoint with plants and millipedes. 
Folia Primatologica, 74. 158, 2003. 

5. X. Valderrama, J. G. Robinson, A. B. Attygalle, and T. Eisner. Seasonal anointment with millipedes in 
a wild primate: A chemical defense against insects?, Journal of Chemical Ecology, 26, 2781, 2000. 

6. P. J. Weldon, J. R. Aldrich, J. A. Klun, J. E. Oliver, and M. Debboun, Benzoquinones from millipedes 
deter mosquitoes and elicit self-anointing in capuchin monkeys ( Cebus spp.), Naturwissenschaften, 
90, 301, 2003. 

7. J. F. Carroll, M. Kramer, P. J. Weldon, and R. G. Robbins, Anointing chemicals and ectoparasites: 
Effects of benzoquinones from millipedes on the lone star tick Amblyomma americanum. Journal of 
Chemical Ecology, 31, 63, 2005. 

8. K. C. Parkes. P. J. Weldon, and R. L. Hoffman, Polydesmidan millipede used in self-anointing by a 
strong-billed woodcreeper ( Xiphocolaptes promeropirhyncus ) from Belize, Ornitologia Neotropica, 
14, 285, 2003. 


2006 by Taylor & Francis Group, LLC 



18 


Insect Repellents: Principles, Methods, and Uses 


9. P. J. Weldon, Defensive anointing: Extended chemical phenotype and unorthodox ecology, 
Chemoecology, 14, 1, 2004. 

10. Herodotus, The Histories, London: Penguin, 1996 (reprint). 

11. D. Charlwood, Did Herodotus describe the first airborne use of mosquito repellents?, Trends in 
Parasitology, 19, 555, 2003. 

12. R. Romi, G. Sabatinelli, and G. Majori, Could malaria reappear in Italy?, Emerging Infectious Disease, 
7, 915, 2001. 

13. T. Owen, Geoponika: Agricultural Pursuits, http://www.ancientlibrary.com/geoponica/index.html, 
1805. 

14. M. A. Braks, R. A. Anderson, and B. G. Knols, Infochemicals in mosquito host selection: Human skin 
microflora and Plasmodium parasites, Parasitology Today, 15, 409, 1999. 

15. R. de Jong and B. G. J. Knols, Selection of biting sites on man by two malaria mosquito species, 
Experientia, 51, 80, 1995. 

16. Y. L. Konan, M. S. Sylla, J. M. Doannio, and S. Traore, Comparison of the effect of two excipients 
(karite nut butter and vaseline) on the efficacy of Cocos nucifera, Elaeis guineensis and Carapa 
procera oil-based repellents formulations against mosquitoes biting in Ivory Coast, Parasite, 10, 181, 
2003. 

17. D. M. S. S. Babu (Ed.), Yoga Ratnakara Visama Jvara Cikitsa-1, in Yoga Ratnakara: Vol. I — 
Purvardham (Sanskrit text with English translation and explanatory notes), Chowkhamba: Sanskrit 
Series Office, p. 698, 2005. 

18. M. J. Bockarie et al., The effect of woodsmoke on the feeding and resting behavior of Anopheles 
gambiae s.s, Acta Tropica, 57, 337, 1994. 

19. E. E. Davis and M. F. Bowen, Sensory physiological basis for attraction in mosquitoes, Journal of the 
American Mosquito Control Association, 10, 316, 1994. 

20. D. E. Moerman, Native American Ethnobotany, Portland, OR: Timber Press, 1998. 

21. K. Kinney, Black Swamp once ruled the land and people, in Sentinel Tribune, Bowling Green, OH, 
1999. 

22. K. T. Silva, Ayurveda, malaria and the indigenous herbal tradition in Sri Lanka, Social Science and 
Medicine, 33, 153, 1991. 

23. K. Palsson and T. G. Jaenson, Comparison of plant products and pyrethroid-treated bed nets for 
protection against mosquitoes (Diptera: Culicidae) in Guinea Bissau, West Africa, Journal of Medical 
Entomology, 36, 144, 1999. 

24. J. M. Dulhunty, K. Yohannes, C. Kourleoutov, V. T. Manuopangai, M. K. Polyn, W. J. Parks, and 
J. H. Bryan, Malaria control in central Malaita, Solomon Islands 2. Local perceptions of the disease 
and practices for its treatment and prevention, Acta Tropica, 75, 185, 2000. 

25. A. D. Rodriguez, R. P. Penilla, M. Henry-Rodriguez, J. Hemingway, A. Francisco Betanzos, and 
J. E. Hernandez-Avila, Knowledge and beliefs about malaria transmission and practices for vector 
control in southern Mexico, Salud Publica de Mexico, 45, 110, 2003. 

26. R. E. Klein, S. C. Weller, R. Zeissig, F. O. Richards, and T. K. Ruebush II, Knowledge, beliefs, and 
practices in relation to malaria transmission and vector control in Guatemala, American Journal of 
Tropical Medicine and Hygiene, 52, 383, 1995. 

27. P. T. Tin-Oo, Pe-Thet-Htoon, Khin-Thet-Wai, W. Parks, and J. Bryan, Gender, mosquitos and 
malaria: Implications for community development programs in Laputta, Myanmar, Southeast Asian 
Journal of Tropica! Medicine and Public Health, 32, 588, 2001. 

28. S.J. Moore, A methodology for developing plant-based products for use against Anopheles 
mosquitoes, Ph.D Thesis, London: London School of Hygiene and Tropical Medicine, 2005. 

29. A. Seyoum, G. F. Killeen, E. W. Kabiru, B. G. Knols, and A. Hassanali, Field efficacy of thermally 
expelled or live potted repellent plants against African malaria vectors in western Kenya, Tropical 
Medicine and International Health, 8, 1005, 2003. 

30. S. W. Lindsay, J. A. Ewald, Y. Samung, C. Apiwathnasom, and F. Nosten, Thanaka ( Limonia 
acidissima) and deet (di-methyl benzamide) mixture as a mosquito repellent for use by Karen Women, 
Medical and Veterinary Entomology, 12, 295, 1998. 


© 2006 by Taylor & Francis Group, LLC 


History of Insect Repellents 


19 


31. R. Vernede, M. M. van Meer. and M. P. Alpers, Smoke as a form of personal protection against 
mosquitoes, a field study in Papua New Guinea, Southeast Asian Journal of Tropical Medicine and 
Public Health, 25, 771, 1994. 

32. J. Stevenson, Zooprophylactic effect of goat herds on human exposure to Phlebotomus martini 
sandflies in East Africa, Transactions of the Royal Society of Tropical Medicine and Hygiene, 
submitted for publication, 2005. 

33. E. J. Hoffman and J. R. Miller, Reduction of mosquito (Diptera: Culicidae) attacks on a human subject 
by combination of wind and vapor-phase DEBT repellent, Journal of Medical Entomology, 39, 935, 
2002 . 

34. K. R. Smith and S. Mehta, The burden of disease from indoor air pollution in developing 
countries: Comparison of estimates, International Journal of Hygiene and Environmental Health, 
206, 279, 2003. 

35. R. A. Cloyd, Natural indeed: Are natural insecticides safer better than conventional insecticides?, 
Illinois Pesticide Review, 17, 2004 (http://www.pesticidesafety.uiuc.edu/newsletter/html/vl7n304. 
pdf). 

36. R. Eisner, ‘Natural’ insecticide research: Still working out the bugs. The Scientist, 5, 14, 1991. 

37. J. E. Casida and G. B. Quistad, Pyrethrum Flowers: Production, Chemistry, Toxicology and Uses, 
New York: Oxford University Press, Inc., 1995. 

38. H. Sangat-Roemantyo, Ethnobotany of Javanese incense, Economic Botany, 44, 413, 1990. 

39. M. Uemura, Eiichiro Ueyama: Developing and promoting insecticide together with pyrethrum, Osaka 
Business Update, 4, 2004 (http://www.ibo.or.jp/e/2004_4/01_l/l_l.html). 

40. WHO, Draft guideline specifications for household insecticide products—mosquito coils, vaporising 
mats, liquid vaporisers, aerosols. Report of the WHO Informal Consultation, Geneva: World Health 
Organization, February 3-6, 1998. 

41. M. S. Mulla, U. Thavara, A. Tawatsin, W. Kong-Ngamsuk, and J. Chompoosri, Mosquito burden and 
impact on the poor: Measures and costs for personal protection in some communities in Thailand, 
Journal of the American Mosquito Control Association, 17, 153, 2001. 

42. M. Ettling, D. A. McFarland, L. J. Schultz, and L. Chitsulo, Economic impact of malaria in Malawian 
households, Tropical Medicine and Parasitology, 45, 74, 1994. 

43. C. E. Lawrance and A. M. Croft, Do mosquito coils prevent malaria? A systematic review of trials. 
Journal of Travel Medicine, 11, 92, 2004. 

44. T. Narahashi, Neuroreceptors and ion channels as the basis for drug action: Past, present, and future, 
Journal of Pharmacology and Experimental Therapeutics, 294, 1, 2000. 

45. D. R. Roberts, T. Chareonviriyaphap, H. H. Harlan, and P. Hshieh, Methods of testing and analyzing 
excito-repellency responses of malaria vectors to insecticides. Journal of the American Mosquito 
Control Association, 13, 13, 1997. 

46. M. Elliott, N. F. Janes, and C. Potter, The future of pyrethroids in insect control, Annual Review of 
Entomology, 23, 443, 1978. 

47. M. Elliott, Properties and applications of pyrethroids. Environmental Health Perspectives, 14, 1, 1976. 

48. D. M. Soderlund and J. R. Bloomquist, Neurotoxic actions of pyrethroid insecticides. Annual Review of 
Entomology, 34, 77, 1989. 

49. P. R. Chadwick, The activity of some pyrethroids, DDT and lindane in smoke from coils for biting 
inhibition, knockdown and kill of mosquitoes (Diptera, Culicidae), Bulletin of Entomological Research, 
65, 97, 1975. 

50. S. P. Frances, Effectiveness of DEET and permethrin, alone, and in a soap formulation as skin and 
clothing protectants against mosquitoes in Australia. Journal of the American Mosquito Control 
Association, 3, 648, 1987. 

51. M. Rowland, T. Freeman, G. Downey, A. Hadi, and M. Saeed, DEET mosquito repellent sold through 
social marketing provides personal protection against malaria in an area of all-night mosquito biting and 
partial coverage of insecticide-treated nets: A case-control study of effectiveness. Tropica! Medicine 
and International Health, 9, 343, 2004. 

52. G. Coveil, Anti-mosquito measures with special reference to India, Health Bulletin, 11, pp. 1-60, 
1943. 


© 2006 by Taylor & Francis Group, LLC 


20 


Insect Repellents: Principles, Methods, and Uses 


53. F. H. Mowrey, Statistics of malaria, in Internal Medicine in World War II, Volume II: Infectious 
Diseases, L. D. Fleaton, J. B. Coates, and W. P. Havens (Eds.), Washington, DC: Medical 
Department, United States Army, pp. 449-463, 1963, http://history.amedd.army.mil/booksdocs/ 
wwii/infectiousdisvolii/frameindex.html. 

54. R. W. Bunn. K. L. Knoght, and W. J. Lacasse, The role of entomology in the preventive medicine 
program of the Armed Forces, Military Medicine, 116, 119, 1955. 

55. C. J. D. Zarafonetis and M. P. Baker, Scrub typhus, in Internal Medicine in World War II, Volume 
II: Infectious Diseases, L. D. Heaton, J. B. Coates, and W. P. Havens (Eds.), Washington, DC: 
Medical Department, United States Army, pp. 111-142, 1963, http://history.amedd.army.mil/ 
booksdocs/wwii/infectiousdisvolii/frameindex.html. 

56. T. F. Whayne, Clothing, in Personal Health Measures and Immunization, J. B. Coates, E. C. Hoff, 
and P. M. Hoff, (Eds.), Office of the Surgeon General Department of the Army, Washington, DC, 
pp. 31-83, 1955, http://history.amedd.army.mil/booksdocs/wwii/PrsnlHlthMsrs/frameindx.html. 

57. U.S. Department of Agriculture, Results of Screening Tests with Materials Evaluated as 
Insecticides, Miticides and Repellents at the Orlando. Orlando: Florida Laboratory E-733, 
Bureau of Entomology and Plant Quarantine United States Department of Agriculture, 1947. 

58. C. Peterson and J. Coats, Insect repellents—past, present and future. Pesticide Outlook, 12, 154, 2001. 

59. A. H. Madden, A. W. Lindquist, and E. F. Knipling, Tests of repellents against chiggers, Journal of 
Economic Entomology, 37, 283. 1944. 

60. R. N. McCulloch, Studies in the control of scrub typhus, Medical Journal of Australia, 1,717, 1946. 

61. R. N. McCulloch and D. F. Waterhouse, Laboratory and field tests of mosquito repellents, in CSIR 
Bulletin No. 213, Melbourne: Australian Government Printer, 1947, p. 1. 

62. Fisher Scientific, Material safety data sheet: 2-ethyl-1,3-hexanediol, mixture of isomers, 99%, in http:// 
fscimage.fishersci.com/msds/71746.htm, 1998. 

63. T. L. Neeper-Bradley, L. C. Fisher, B. L. Butler, and B. Ballantyne, Evaluation of the developmental 
toxicity potential of 2-ethyl-1,3-hexanediol in the rat by cutaneous application, Journal of Toxicology: 
Cutaneous and Ocular Toxicology, 13, 203, 1994. 

64. E. T. McCabe, W. F. Barthel, S. I. Gertler, and S. A. Hall, Insect repellents, III, /V,/V-diethylamides, 
Journal of Organic Chemistry, 19, 493, 1954. 

65. J. R. Roberts and I. R. Reigart, Does anything beat DEET?, Pediatric Annual, 33, 443, 2004. 

66. C. E. Schreck and T. P. McGovern, Repellents and other personal protection strategies against Aedes 
albopictus. Journal of the American Mosquito Control Association, 5, 247, 1989. 

67. M. Debboun, D. Strickman, V. B. Solberg, R. C. Wilkerson, K. R. McPherson, C. Golenda, L. Keep, 
R. A. Wirtz, R. Burge, and T. A. Klein. Field evaluation of DEET and a piperidine repellent against 
Aedes communis (Diptera: Culicidae) and Simulium venustum (Diptera: Simuliidae) in the Adirondack 
Mountains of New York, Journal of Medical Entomology, 37, 919, 2000. 

68. D. R. Barnard, U. R. Bernier. K. H. Posey, and R. D. Xue, Repellency of IR3535, KBR3023, para- 
menthane-3,8-diol, and DEET to black salt marsh mosquitoes (Diptera: Culicidae) in the Everglades 
National Park. Journal of Medical Entomology, 39, 895, 2002. 

69. J. E. Cilek, I. L. Petersen, and C. E. Hallmon, Comparative efficacy of IR3535 and DEET as repellents 
against adult Aedes aegypti and Culex quinquefasciatus. Journal of the American Mosquito Control 
Association, 20, 299, 2004. 

70. R. K. Gupta and L. C. Rutledge, Controlled release repellent formulations on human volunteers under 
three climatic regimens, Journal of the American Mosquito Control Association, 7, 490, 1991. 

71. Y. Trongtokit, Y. Rongsriyam, N. Komalamisra, P. Krisadaphong, and C. Apiwathnasorn, Laboratory 
and field trial of developing medicinal local Thai plant products against four species of mosquito 
vectors. Southeast Asian Journal of Tropical Medicine and Public Health, 35, 325, 2004. 

72. B. Annis, Comparison of the effectiveness of two DEET formulations against Aedes albopictus in the 
Philippines, Journal of the American Mosquito Control Association, 7, 543, 1991. 

73. D. R. Barnard and R. D. Xue, Laboratory evaluation of mosquito repellents against Aedes albopictus, 
Culex nigripalpus, and Ochierotatus triseriatus (Diptera: Culicidae), Journal of Medical Entomology, 
41, 726, 2004. 


© 2006 by Taylor & Francis Group, LLC 


History of Insect Repellents 


21 


74. C. F. Curtis, J. D. Lines, J. Ijumba, A. Callaghan, N. Hill, and M. A. Karimzad, The relative 
efficacy of repellents against mosquito vectors of disease. Medical and Veterinary Entomology , 1, 
109, 1987. 

75. R. E. Coleman, A. L. Richards, G. J. Magnon, C. S. Maxwell, M. Debboun, T. A. Klein, and 
R. A. Wirtz, Laboratory and held trials of four repellents with Culex pipiens (Diptera: Culicidae), 
Journal of Medical Entomology, 31, 17, 1994. 

76. S. P. Frances, N. Van Dung, N. W. Beebe, and M. Debbboun, Field evaluation of repellent 
formulations containing DEET and picaridin against mosquitoes in Northern Territory, Australia, 
Journal of Medical Entomology, 41, 414, 2004. 

77. G. L. Chiang, S. L. Tay, K. L. Eng, and S. T. Chan, Effectiveness of repellent/insecticidal bars against 
malaria and Hlariasis vectors in peninsular Malaysia, Southeast Asian Journal of Tropical Medicine 
and Public Health, 21, 412, 1990. 

78. C. Costantini, A. Badolo, and E. Ilboudo-Sanogo, Field evaluation of the efficacy and persistence of 
insect repellents DEET, IR3535. and KBR 3023 against Anopheles gambiae complex and other 
Afrotropical vector mosquitoes. Transactions of the Royal Society of Tropical Medicine and Hygiene, 
98, 644, 2004. 

79. J. K. Trigg. Evaluation of a eucalyptus-based repellent against Anopheles spp. in Tanzania, Journal of 
the American Mosquito Control Association, 12, 243, 1996. 

80. T. W. Walker, L. L. Robert, R. A. Copeland, A. K. Githeko. R. A. Wirtz, J. I. Githure, and 
T. A. Klein, Field evaluation of arthropod repellents, DEET and a piperidine compound, AI3-37220, 
against Anopheles funestus and Anopheles arabiensis in western Kenya, Journal of the American 
Mosquito Control Association, 12, 172, 1996. 

81. J. Govere, D. N. Durrheim, L. Baker, R. Hunt, and M. Coetzee. Efficacy of three insect repellents 
against the malaria vector Anopheles arabiensis. Medical and Veterinary Entomology, 14, 441, 2000. 

82. J. Govere, L. E. Braack, D. N. Durrheim, R. H. Hunt, and M. Coetzee, Repellent effects on Anopheles 
arabiensis biting humans in Kruger Park, South Africa, Medical and Veterinary Entomology, 15, 287, 
2001 . 

83. S. P. Frances, N. Eikarat, B. Sripongsai, and C. Eamsila, Response of Anopheles dims and Aedes 
albopictus to repellents in the laboratory. Journal of the American Mosquito Control Association, 9, 
474, 1993. 

84. S. P. Frances, T. A. Klein, D. W. Hildebrandt, R. Burge, C. Noigamol, N. Eikarat, B. Sripongsai, and 
R. A. Wirtz, Laboratory and held evaluation of DEET, CIC-4, and AI3-37220 against Anopheles 
dims (Diptera: Culicidae) in Thailand, Journal of Medical Entomology, 33, 511, 1996. 

85. S. J. Moore, A. Lenglet, and N. Hill, Field evaluation of three plant-based insect repellents against 
malaria vectors in Vaca Diez Province, the Bolivian Amazon. Journal of the American Mosquito 
Control Association, 18, 107, 2002. 

86. S. P. Frances, R. D. Cooper, S. Popat, and A. W. Sweeney, Field evaluation of the repellents DEET, 
CIC-4, and AI3-37220 against Anopheles in Lae, Papua New Guinea, Journal of the American 
Mosquito Control Association, 15, 339, 1999. 

87. S. P. Frances, N. Van Dung, N. W. Beebe, and M. Debboun, Field evaluation of repellent 
formulations against daytime and nighttime biting mosquitoes in a tropical rainforest in northern 
Australia, Journal of Medical Entomology, 39, 541, 2002. 

88. R. E. Coleman, L. L. Robert, L. W. Roberts, J. A. Glass, D. C. Seeley, A. Laughinghouse, 
P. V. Perkins, and R. A. Wirtz, Laboratory evaluation of repellents against four anopheline 
mosquitoes (Diptera: Culicidae) and two phlebotomine sand hies (Diptera: Psychodidae), Journal 
of Medical Entomology, 30, 499, 1993. 

89. B. Alexander, H. Cadena, M. C. Usma, and C. A. Rojas, Laboratory and held evaluations of a 
repellent soap containing diethyl toluamide (DEET) and permethrin against phlebotomine sand hies 
(Diptera: Psychodidae) in Valle del Cauca, Colombia, American Journal of Tropica! Medicine and 
Hygiene, 52, 169, 1995. 

90. M. Kalyanasundaram, R. Srinivasan, S. Subramanian, and K. N. Panicker, Relative potency of DEPA 
as a repellent against the sand fly Phlebotomus papatasi. Medical and Veterinary Entomology, 8, 68, 
1994. 


2006 by Taylor & Francis Group, LLC 



22 


Insect Repellents: Principles, Methods, and Uses 


91. L. L. Robert, R. E. Coleman, D. A. Lapointe, P. J. Martin. R. Kelly, and J. D. Edman, Laboratory and 
field evaluation of five repellents against the black files Prosimulium mixtum and P. fuscum (Diptera: 
Simuliidae ), Journal of Medical Entomology, 29, 267, 1992. 

92. T. M. Ho and M. K. Fauziah, Laboratory evaluation of two commercial repellants against 
Leptotrombidium fletcheri (Acari: Trombiculidae), Southeast Asian Journal of Tropica! Medicine 
and Public Health, 24, 165, 1993. 

93. S. P. Frances, Response of a chigger, Eutrombicula histri (Acari: Trombiculidae) to repellent and 
toxicant compounds in the laboratory, Journal of Medical Entomology, 31, 628, 1994. 

94. C. Eamsila, S. P. Frances, and D. Strickman, Evaluation of permethrin-treated military uniforms for 
personal protection against malaria in northeastern Thailand, Journal of the American Mosquito 
Control Association, 10, 515, 1994. 

95. S. Kumar, S. Prakash, and K. M. Rao, Comparative activity of three repellents against the ticks 
Rhipicephalus sanguineus and Argas persicus. Medical and Veterinary Entomology, 6, 47, 1992. 

96. V. B. Solberg, T. A. Klein, K. R. McPherson, B. A. Bradford, J. R. Burge, and R. A. Wirtz, Field 
evaluation of DEET and a piperidine repellent (AI3-37220) against Amblyomma americanum (Acari: 
Ixodidae), Journal of Medical Entomology, 32, 870, 1995. 

97. B. Salafsky, Y. X. He, J. Li. T. Shibuya, and K. Ramaswamy, Short report: Study on the efficacy of a 
new long-acting formulation of (VjV-diethyl-m-toluamide (DEET) for the prevention of tick 
attachment, American Journal of Tropical Medicine and Hygiene, 62, 169, 2000. 

98. A. M. Pretorius, M. Jensenius, F. Clarke, and S. H. Ringertz, Repellent efficacy of DEET and KBR 3023 
against Amblyomma hebraeum (Acari: Ixodidae), Journal of Medical Entomology, 40, 245, 2003. 

99. S. Kumar, S. Prakash, and K. M. Rao, Comparative activity of three repellents against bedbugs Cimex 
hemipterus (Fabr.), Indian Journal of Medical Research, 102, 20, 1995. 

100. Z. A. Mehr, L. C. Rutledge, and J. L. Inase, Evaluation of commercial and experimental 
repellents against Xenopsylla cheopis (Siphonaptera: Pulicidae), Journal of Medical Entomology, 
21, 665, 1984. 

101. L. Goodyer and R. H. Behrens, Short report: The safety and toxicity of insect repellents, American 
Journal of Tropical Medicine and Hygiene, 59, 323, 1998. 

102. U.S. Environmental Protecton Agency, Pesticide Registration Standard for (VW-diethyl-m-toluamide 
(DEET), Washington, DC: Office of Pesticides and Toxic Substances Special Pesticides Review 
Division, United States Environmental Protection Agency, 1980. 

103. E. H. Roland, J. E. Jan, and J. M. Rigg, Toxic encephalopathy in a child after brief exposure to insect 
repellents, Canadian Medical Association Journal, 132, 155, 1985. 

104. T. G. Osimitz and J. V. Murphy, Neurological effects associated with use of the insect repellent 
A.iV-diethyl-m-toluamide (DEET), Journal of Toxicology: Clinical Toxicology, 35, 435, 1997. 

105. G. Briassoulis, M. Narlioglou, and T. Hatzis, Toxic encephalopathy associated with use of DEET 
insect repellents: a case analysis of its toxicity in children. Human and Experimental Toxicology, 20, 8, 
2001 . 

106. D. L. Sudakin and W. R. Trevathan, DEET: A review and update of safety and risk in the general 
population. Journal of Toxicology: Clinical Toxicology, 41, 831, 2003. 

107. T. G. Osimitz and R. H. Grothaus, The present safety assessment of DEET, Journal of the American 
Mosquito Control Association, 11, 274, 1995. 

108. M. B. Abou-Donia, K. R. Wilmarth, K. F. Jensen, F. W. Oehme, and T. L. Kurt. Increased 
neurotoxicity following concurrent exposure to pyridostigmine bromide, DEET, and chlorpyrifos. 
Fundamental and Applied Toxicology, 34, 201, 1996. 

109. J. C. Veltri, T. G. Osimitz, D. C. Bradford, and B. C. Page, Retrospective analysis of calls to poison 
control centers resulting from exposure to the insect repellent (VJV-diethyl-m-toluamide (DEET) from 
1985-1989, Journal of Toxicology: Clinical Toxicology, 32, 1, 1994. 

110. M. S. Fradin, Mosquitoes and mosquito repellents: A clinician's guide. Annals of Internal Medicine, 
128, 931, 1998. 

111. U.S. Environmental Protection Agency, Reregistration Eligibility Decision (RED): DEET, EPA738- 
R-98-010, in Prevention, Pesticides and Toxic Substances, Washington, DC: United States 
Environmental Protection Agency, 1998. 


© 2006 by Taylor & Francis Group, LLC 



History of Insect Repellents 


23 


112. R. McGready, K. A. Hamilton. J. A. Simpson, T. Cho, C. Luxemburger, R. Edwards, S. Looareesuwan, 
N. J. White, F. Nosten, and S. W. Linds, Safety of the insect repellent N, IV-diethyl-m-toluamide (DEET) 
in pregnancy, American Journal of Tropica! Medicine and Hygiene , 65, 285, 2001. 

113. M. Rowland, K. A. Hamilton, J. A. Simpson, T. Cho, C. Luxemburger, R. Edwards, 
S. Looareesuwan, N. J. White, F. Nosten, and S. W. Lindsay, DEET mosquito repellent provides 
personal protection against malaria: A household randomized trial in an Afghan refugee camp in 
Pakistan, Tropical Medicine and International Health , 9, 335, 2004b. 

114. S. S. Rao, S. Prakash, S. Kumar, U. Kaveeshwar, B. K. Bhattacharya, D. K. Jaiswal, and 
K. M. Rao, Toxicologic studies of an insect repellent AUV-diethylphenylacetamide, Indian Journal 
of Medical Research, 85, 626, 1987. 

115. S. S. Rao, R. V. Swamy, and P. K. Ramachandran. Toxicity and metabolism of a new insect repellent 
AUV-diethylphenylacetamide in mice, rats and guinea pigs on cutaneous application, Toxicology, 58, 
81, 1989. 

116. C. Sadanandane, N. Mathew, P. Jambulingam, and K. Kalyanasundaram, Laboratory and field 
evaluation of controlled release formulation of the insect repellents /V,/V-diethyl-m-toluamide (DEET) 
and lV,/V-diethyl phenylacetamide (DEPA) against mosquito vectors, Indian Journal of Medical 
Research, 113, 108, 2001. 

117. S. Kumar, S. Prakash, R. K. Sharma, S. K. Jain, M. Kalyanasundaram, R. V. Swamy, and K. M. Rao, 
Field evaluation of three repellents against mosquitoes, blackflies and land leeches, Indian Journal of 
Medical Research, 80, 541, 1984. 

118. M. Debboun and J. Wagman, In vitro repellency of /V,/V-diethyl-3-methylbenzamide and AL/V-diethyl- 
phenylacetamide analogs against Aedes aegypti and Anopheles stephensi (Diptera: Culicidae), 
Journal of Medical Entomology, 41, 430, 2004. 

119. S. S. Rao and K. M. Rao, Insect repellent /V,/V-diethylphenylacetamide: An update. Journal of 
Medical Entomology, 28, 303, 1991. 

120. DRDO. Multi-insect repellent diethylphenylacetamide spray, Technology Focus, 12, 2004 (http:// 
www.drdo.com/pub/techfocus/apr04/welcometfapril04.htm). 

121. E. Worrall, S. Basu, and K. Hanson, The relationship between socio-economic status and malaria: A 
review of the literature, in Ensuring that Malaria Control Interventions Reach the Poor, 
London: The World Bank, p. 46, 2002, http://siteresources.worldbank.org/INTMALARIA/ 
Resources/SESMalariaBackgroundPaper.pdf. 

122. M. Klier and F. Kuhlow, Neue Insektenabwehrmittel—am Stickstoff disubstituierte beta-Alaninder- 
ivate, Journal of the Society of Cosmetic Chemists, 27, 141, 1976. 

123. WHO, IR 3535 ethyl butylacetylaminoproprionate information: Interim specification WHO/IS/TC/ 
667/2001, Geneva: World Health Organization (http://www.who.int/whopes/quality/en/Icaridin_ 
spec_eval_0ct_2004.pdf). 

124. EPA, 3-[/V-Butyl-/V-acetyl]-aminopropionic acid, Ethyl Ester (113509) Technical Document, 
Washington, DC: Environmental Protection Agency, Biopesticides and Pollution Prevention Division 
1999, (http://www.epa.gov/oppbppdl/biopesticides/ingredients/tech_docs/techl 13509.htm). 

125. U. Thavara, A. Tawatsin, J. Chompoosri, W. Suwonkerd, U. R. Chansang, and P. Asavadachanukorn, 
Laboratory and field evaluations of the insect repellent 3535 (ethyl butylacetylaminopropionate) and 
DEET against mosquito vectors in Thailand, Journal of the American Mosquito Control Association, 17, 
190, 2001. 

126. H. H. Yap, Laboratory Repellent Test on IR3535 and DEET at 25% w/v Formulations Against Aedes 
Albopictus and Anopheles Maculatus, Penang, Malaysia: Universiti Sains Malaysia, Vector Control 
Research Unit, 1998a. 

127. H. H. Yap, Field Efficacy Test on Insect Repellent Samples with IR3535 or DEET as Active Ingredients 
Against Aedes Albopictus and Culex quinquefaciatus in the tropical environment, Penang, Malaysia: 
Universiti Sains Malaysia, Vector Control Research, 1998b. 

128. F. Rettich, Laboratory and field evaluation of two new mosquito repellents, in 13th European SOVE 
Meeting, Ankara: European Society of Vector Ecology, 1999. 

129. B. W. Kruger et al.. Agents for repelling insects and mites European Patent, 2.3, Bayer AG, 281, 903, 
1998. 


© 2006 by Taylor & Francis Group, LLC 


24 


Insect Repellents: Principles, Methods, and Uses 


130. EPA, New pesticide fact sheet picaridin, Washington, DC: Prevention, Pesticides and Toxic 
Substances, United States Environmental Protection Agency 2005, (http://www.epa.gov/opprd001/ 
factsheets/picaridin.pdf). 

131. G. Nentwig, J. Boeckh, F. P. Hoever, B. W. Kruger, K. Roder, and A. G. Bayer, Bayrepel (KBR3023), a 
new mosquito repellent: from laboratory synthesis to a worldwide commercial product, in 3rd European 
Conference on Travel Medicine , Florence: World Health Organization Collaborating Centre for Travel 
Medicine, 2002. 

132. S. P. Frances and R. D. Cooper, Personal protection measures against mosquitoes: A brief history and 
current use of repellents by the Australian Defence Forces, ADF Health, 3, 58, 2002. 

133. Z. A. Mehr, L. C. Rutledge, E. L. Morales, V. E. Meixsell, and D. W. Korte, Laboratory evaluation of 
controlled-release insect repellent formulations, Journal of the American Mosquito Control Associ¬ 
ation, 1, 143, 1985. 

134. J. A. Klun, A. Khrimian, A. Margaryan, M. Kramer, and M. Debboun, Synthesis and repellent efficacy of 
a new chiral piperidine analog: Comparison with DEET and Bayrepel activity in human-volunteer 
laboratory assays against Aedes aegypti and Anopheles stephensi. Journal of Medical Entomology, 40, 
293, 2003. 

135. H. H. Yap, K. Jahangir, A. S. Chong, C. R. Adanan, N. L. Chong, Y. A. Malik, and B. Rohaizat, Field 
efficacy of a new repellent. KBR 3023, against Aedes albopictus (Skuse) and Culex quinquefasciatus 
(Say) in a tropical environment, Journal of Vector Ecology, 23, 62, 1998. 

136. A. Badolo, E. Ilboudo-Sanogo, A. P. Ouedraogo, and C. Costantini, Evaluation of the sensitivity of 
Aedes aegypti and Anopheles gambiae complex mosquitoes to two insect repellents: DEET and KBR 
3023, Tropical Medicine and International Health, 9, 330, 2004. 

137. WHO, Report of the 4th WHOPES Working Group meeting 4-5th December 2000: Review of IR3535, 
KBR3023, (RS)-methoprene 20% EC, pyriproxyfen 0.5% GR and lambda-cyhalothrin 2.5% CS. WHO/ 
CDS/WHOPES/2001.2, Geneva: World Health Organization, p. 102, 2001. 

138. CDC, Updated Information regarding mosquito repellents: April 22,2005, http://www.cdc.gov/elcosh/ 
docs/d0600/d000695/d000695.pdf, Atlanta: Centers for Disease Control and Prevention, 2005. 

139. T. P. McGovern, C. E. Schreck, and A. Jackson, Mosquito repellents: Alicyclic amides as repellents for 
Aedes aegypti and Anopheles quadrimaculatus. Mosquito News, 38, 346, 1978. 

140. L. L. Robert, R. E. Coleman, D. A. Lapointe, P. J. Martin, R. Kelly, and J. D. Edman, Laboratory and 
field evaluation of five repellents against the black flies Prosimulium mixtum and P. fuscum (Diptera: 
Simuliidae), Journal of Medical Entomology, 29, 267, 1992. 

141. M. Debboun, D. Strickman, T. A. Klein, J. A. Glass, E. Wylie, A. Laughinghouse, R. A. Wirtz, and 
R. K. Gupta, Laboratory evaluation of AI3-37220, AI3-35765, CIC-4, and DEET repellents against 
three species of mosquitoes, Journal of the American Mosquito Control Association, 15, 342, 1999. 

142. J. A. Klun, D. Ma, and R. Gupta, Optically active arthropod repellents for use against disease vectors. 
Journal of Medical Entomology, 37, 182, 2000. 

143. S. P. Frances, R. D. Cooper, and A. W. Sweeney, Laboratory and field evaluation of the repellents 
DEET, CIC-4, and AI3-37220 against Anopheles farauti (Diptera: Culicidae) in Australia, Journal of 
Medical Entomology, 35, 690, 1998. 

144. S. P. Frances, T. A. Klein, D. W. Hildebrandt, R. Burge, C. Noigamol, N. Eikarat. B. Sripongsai, and 
R. A. Wirtz, Laboratory and field evaluation of deet, CIC-4, and AI3-37220 against Anopheles dirus 
{Diptera: Culicidae) in Thailand, Journal of Medical Entomology, 33, 511, 1996. 

145. M. J. Perich, D. Strickman, R. A. Wirtz, S. A. Stockwell, J. I. Glick, R. Burge, G. Hunt, andP. G. Lawyer, 
Field evaluation of four repellents against Leptoconops americanus (Diptera: Ceratopogonidae) biting 
midges, Journal of Medical Entomology, 32, 306, 1995. 

146. H. L. Snodgrass, J. T. Houpt, and P. A. Beall (Eds.), Dermal Penetration of the Candidate Insect 
Repellent AI3-37220 in Swine and Rabbits, Baltimore, MD: U.S. Army Center for Health Promotion and 
Preventative Medicine, Aberdeen Proving Ground, 1991. 

147. H. L. Snodgrass and H. G. Harvey, Toxicology Study No. 86-8104-98, The Acute Toxicity of the 
Diastereomers of the Insect Repellent AI3-37220, Baltimore, MD: U.S. Army Center for Health 
Promotion and Preventative Medicine, Aberdeen Proving Ground, MD, 1998. 


© 2006 by Taylor & Francis Group, LLC 


History of Insect Repellents 


25 


148. J. A. Klun, W. F. Schmidt, and M. Debboun, Stereochemical effects in an insect repellent, Journal of 
Medical Entomology, 38, 809, 2001. 

149. J. F. Carroll, J. A. Klun, and M. Debboun, Repellency of DEET and SS220 applied to skin involves 
olfactory sensing by two species of ticks. Medical and Veterinary Entomology, 19, 101, 2005. 

150. J. A. Klun, D. Strickman, E. Rowton, J. Williams, M. Kramer, D. Roberts, and M. Debboun, 
Comparative resistance of Anopheles albimanus and Aedes aegypti to /V,,/V-diethyl-3-methylbenzamide 
(DEET) and 2-methylpiperidinyl-3-cyclohexen- 1-carboxamide (AI3-37220) in laboratory human- 
volunteer repellent assays. Journal of Medical Entomology, 41,418, 2004. 

151. J. Core, R. Bliss, and A. Flores, ARS partners with defense department to protect troops from insect 
vectors, Agricultural Research, 25, 2005, (http://www.afpmb.org/bulletin/vol25/vector0905.pdf). 

152. J. T. Houpt, Toxicology Study No. 85-XC-3929-02, Protocol No. 3929-13b-01-04-01, The Skin 
Sensitization of the Repellent SS-220 in Guinea Pigs, Baltimore, MD: U.S. Army Center for Health 
Promotion and Preventative Medicine, Aberdeen Proving Ground, 2002. 

153. H. L. Snodgrass and J. T. Houpt, Toxicology Study No. 85-XC-3929-02, Protocol No. 3929-23-01-07- 
02, The Acute Dermal Toxicity of the Repellent SS-220 in Guinea Pigs, Baltimore, MD: U.S. Army 
Center for Health Promotion and Preventative Medicine, Aberdeen Proving Ground, 2002. 

154. H. L. Snodgrass, Toxicology Study No. 85-XC-3929-02. Protocol No. 3929-10-99-06-01, The Primary 
Eye Irritation of the Repellent SS-220 in Rabbits, Baltimore, MD: U.S. Army Center for Health 
Promotion and Preventative Medicine, Aberdeen Proving Ground, 2002. 

155. H. L. Snodgrass, Toxicology Study No. 85-XC-3929-02. Protocol No. 3929-11-01-07-04, The Primary 
Skin Irritation of the Repellent SS-220 in Rabbits, Baltimore, MD: U.S. Army Center for Health 
Promotion and Preventative Medicine, Aberdeen Proving Ground, 2002. 

156. H. L. Snodgrass, Toxicology Study No. 85-XC-3929-02, Protocol No. 3929-18-01-07-01, The Acute 
Oral Toxicity of the Repellent SS-220 in Rats, Baltimore, MD: U.S. Army Center for Health Promotion 
and Preventative Medicine, Aberdeen Proving Ground, 2002. 

157. Leahy, S., Building a better bug beater, in Wired News, http://www.wired.com/news/medtech/ 
0,1286,59363,00.html, 2003. 

158. L. Garson and M. E. Winnike, Relationship between insect repellency and physical parameters— 
a review, Journal of Medical Entomology, 5, 339, 1968. 

159. M. L. Gabel, I. S. Spencer, and W. A. Akers, Evaporation rates and protection times of repellents. 
Mosquito News, 36, 141, 1976. 

160. A. A. Khan, H. I. Maibach, and D. L. Skidmore, A study of insect repellents. 2. Effect of temperature on 
protection time. Journal of Economic Entomology, 66, 437, 1973. 

161. P. V. Wood, The effect of ambient humidity on the repellency of ethylhexanediol to Aedes aegypti, 
Canadian Entomologist, 100, 1331, 1968. 

162. A. Kroeger, A. Gerhardus, G. Kruger, M. Mancheno, and K. Pesse, The contribution of repellent soap to 
malaria control, American Journal of Tropical Medicine and Hygiene, 56, 580, 1997. 

163. L. M. Rueda, L. C. Rutledge, and R. K. Gupta, Effect of skin abrasions on the efficacy of the repellent 
DEET against Aedes aegypti. Journal of the American Mosquito Control Association, 14, 178, 1998. 

164. T. S. Spencer, R. K. Shimmin, S. Bayles, and W. A. Akers, Consideration of repellent screening 
standards report no. 20, U.S. Research and Development Command, San Francisco: Letterman Army 
Institute, p. 23, 1974. 

165. Y. Trongtokit, Y. Rongsriyam, N. Komalamisra, and C. Apiwathnasom, Comparative repellency of 38 
essential oils against mosquito bites. Phytotherapy Research, 19, 303, 2005. 

166. Terpine Consortium, Test Plan for Terpenoid Primary Alcohols and Related Esters, Washington, DC: 
The Flavor and Fragrance High Production Volume Consortia, http://www.epa.gov/chemrtk/terprial/ 
cl2965tp.pdf, 1999. 

167. ATSDR, Chemical technical summary for public health and public safety professionals, Atlanta, GA: 
Agency for Toxic Substances and Disease Registry, http://www.atsdr.cdc.gov/consultations/DEET/ 
index.html, 2005. 

168. R. J. Feldmann and H. I. Maibach, Absorption of some organic compounds through the skin in man. 
Journal of Investigative Dermatology, 54, 399, 1970. 

169. C. G. MacNay, Studies repellents for biting flies, Canadian Entomologist, 71, 38, 1939. 


© 2006 by Taylor & Francis Group, LLC 


26 


Insect Repellents: Principles, Methods, and Uses 


170. A. J. S. Weaving and N. K. Sylvester, Pyrethrum as an insect repellent. Part II: A laboratory technique for 
its evaluation as a mosquito repellent, and the influence of formulation on persistence introduction, 
Pyrethrum Post, 9, 31, 1967. 

171. L. C. Rutledge, R. K. Gupta, Z. A. Mehr, M. D. Buescher, and W. G. Reifenrath, Evaluation of 
controlled-release mosquito repellent formulations, Journal of the American Mosquito Control 
Association, 12, 39, 1996. 

172. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Addition of perfume fixatives to mosquito repellents to 
increase protection time. Mosquito News, 35, 23, 1975. 

173. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Insect repellents: Effect of mosquito and repellent- 
related factors on protection time. Journal of Economic Entomology, 68, 43, 1975. 

174. A. P. Kurtz, J. A. Logan, and W. A. Akers, More Effective Topical Repellents Against Malaria-bearing 
Mosquitoes: Review of Volunteer Tests of Mosquito Repellent Formulations October 1969-September 
1971, San Francisco, CA: Letterman Army Institute of Research, 1973. 

175. AFPM Board (Ed.), Personal Protective Measures Against Insects and Other Arthropods of Military 
Significance Technical Guide 36, Washington, DC: Defense Pest Management Information Analysis 
Center Forest Glen Section, Walter Reed Army Medical Center, 2002. 

176. R. K. Gupta, A. W. Sweeney, L. C. Rutledge, R. D. Cooper, S. P. Frances, and D. R. Westrom, 
Effectiveness of controlled-release personal-use arthropod repellents and permethrin-impregnated 
clothing in the field. Journal of the American Mosquito Control Association, 3, 556, 1987. 

177. B. Annis, Comparison of the effectiveness of two formulations of DEET against Anopheles flavirostris. 
Journal of the American Mosquito Control Association, 6, 430, 1990. 

178. S. P. Frances, R. D. Cooper, S. Popat, and N. W. Beebe, Field evaluation of repellents containing DEET 
and AI3-37220 against Anopheles koliensis in Papua New Guinea, Journal of the American Mosquito 
Control Association, 17, 42, 2001. 

179. A. M. Croft, D. Baker, and M. J. Von Bertele, An evidence based vector control strategy for military 
deployments: The British Army experience, Medicine in the Tropics, 91, 2001. 

180. H. Qui, H. W. lun. M. Dzimianski, and J. McCall, Reduced transdermal absorption of V,V-diethyl-/n- 
toluamide from a new topical insect repellent fomulation, Pharmaceutical Development and Tech¬ 
nology, 2, 33, 1997. 

181. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Increased abrasion and wash resistance of repellents 
with the addition of polymers. Mosquito News, 37, 123, 1977. 

182. I. M. Gambel, J. F. Brundage, R. A. Kuschner, and P. W. Kelley, Deployed US Army soldiers’ 
knowledge and use of personal protection measures to prevent arthropod-related casualties. Journal of 
Travel Medicine, 5, 217, 1998. 

183. S. P. Frances, A. M. Auliff, M. D. Edstein, and R. D. Cooper, Survey of personal protection measures 
against mosquitoes among Australian defense force personnel deployed to East Timor, Military 
Medicine, 168, 227, 2003. 

184. L. R. Lindsay, G. A. Surgeoner, I. D. Heal, and G. J. Gallivan, Evaluation of the efficacy of 3% citronella 
candles and 5% citronella incense for protection against field populations of Aedes mosquitoes. Journal 
of the American Mosquito Control Association, 12, 293, 1996. 

185. I. A. Nolen, R. H. Bedoukan, R. E. Maloney, and D. L. Kline, Method apparatus and compositions for 
inhibiting the human scent tacking ability of mosquitoes in environmentally defined three dimensional 
spaces, in US Patent 6,362,235, Alexandria, VA: US Patent Office, 2002. 

186. D. L. Kline, U. R. Bernier, K. H. Posey, andD. R. Barnard, Olfactometric evaluation of spatial repellents 
for Aedes aegypti. Journal of Medical Entomology, 40, 463, 2003. 

187. B. Alten, U. R. Bernier, K. H. Posey, and D. R. Barnard, Field evaluation of an area 
repellent system (Thermacell) against Phlebotomus papatasi (Diptera: Psychodidae) and Ochlerotatus 
caspius (Diptera: Culicidae) in Sanliurfa Province, Journal of Medical Entomology, Turkey, 40, 930, 
2003. 

188. C. F. Curtis and N. Hill, Comparison of methods of repelling mosquitoes, Entomologia Expermentalis et 
Applicata, 49, 175, 1988. 

189. S. E. Hewitt, M. Farhan, H. Urhaman, N. Muhammad, M. Kamal, and M. W. Rowland, Self-protection 
from malaria vectors in Pakistan: An evaluation of popular existing methods and appropriate new 


© 2006 by Taylor & Francis Group, LLC 



History of Insect Repellents 


27 


techniques in Afghan refugee communities, Annals of Tropical Medicine and Parasitology, 90, 337, 
1996. 

190. U. R. Bernier. K. D. Furman, D. L. Kline, S. A. Allan, and D. R. Barnard, Comparison of contact and 
spatial repellency of catnip oil and AQV-diethyl-3-methylbenzamide (DEET) against mosquitoes. 
Journal of Medical Entomology, 42, 306, 2005. 

191. S. J. Moore and A. Lenglet, An overview of plants as insect repellents, in Traditional Medicine, 
Medicinal Plants and Malaria, G. Bodeker (Ed.), London: Taylor and Francis, pp. 343-363, 2004. 

192. U. R. Bernier, M. M. Booth, andR. A. Yost, Analysis of human skin emanations by gas chromatography/ 
mass spectrometry. 2. Identification of volatile compounds that are candidate attractants for the yellow 
fever mosquito (Aedes aegypti ), Analytical Chemistry, 12, 747, 2000. 

193. U. R. Bernier, D. L. Kline, K. H. Posey, M. M. Yost, and D. R. Barnard, Synergistic attraction of Aedes 
aegypti (L.) to binary blends of L-lactic acid and acetone, dichloromethane, or dimethyl disulfide, 
Journal of Medical Entomology, 40, 653, 2003. 

194. M. F. Bowen, The sensory physiology of host-seeking behavior in mosquitoes. Annual Review of 
Entomology, 36, 139, 1991. 

195. M. F. Bowen, Sensory aspects of host location in mosquitoes, CIBA Foundation Symposia, 200, 197, 
1996. 

196. R. de Jong and B. G. J. Knols, Olfactory responses of host-seeking Anopheles gambiae s.s. Giles 
(Diptera: Culicidae), Ada Tropica, 59, 333, 1995. 

197. A. E. Eiras and P. C. Jepson, Responses of female Aedes aegypti (Diptera: Culicidae) to host odors 
and convection currents using an olfactometer bioassay. Bulletin of Entomological Research, 84, 207, 
1994. 

198. T. P. Flealy, M. J. Copland, A. Cork, A. Przyborowska, and J. M. Halket. Landing responses of 
Anopheles gambiae elicited by oxocarboxylic acids. Medical and Veterinary Entomology, 16, 126, 
2002 . 

199. B. G. Knols, R. de Jong, and W. Takken, Differential attractiveness of isolated humans to mosquitoes in 
Tanzania, Transactions of the Royal Society of Tropical Medicine and Hygiene, 89, 604, 1995. 

200. B. G. Knols, On human odor, malaria mosquitoes, and limburger cheese. Lancet, 348, 1322, 1996. 

201. W. Takken, The role of olfaction in host-seeking of mosquitoes: A review, Insect Science and its 
Applications, 12, 287, 1991. 

202. W. Takken, B. G. J. Knolls, and H. Otten. Interactions between physical and olfactory cues in the host¬ 
seeking behavior of mosquitoes: The role of relative humidity. Annals of Tropical Medicine and 
Parasitology, 91, SI 19, 1997. 

203. C. Costantini. M. A. Birkett, G. Gibson. J. Ziesmann, N. F. Sagnon, H. A. Mohammed, M. Coluzzi. and 
J. A. Pickett, Electroantennogram and behavioral responses of the malaria vector Anopheles gambiae to 
human-specific sweat components. Medical and Veterinary Entomology, 15, 259, 2001. 

204. Y. T. Qiu, R. C. Smallegange, S. Hoppe, J. J. A. Van Loon, E. Bakker, and W. Takken, Behavioral and 
electrophysiological responses of the malaria mosquito Anopheles gambiae Giles sensu stricto (Diptera: 
Culicidae) to human skin emanations. Medical and Veterinary Entomology, 18, 429, 2004. 

205. E. E. Davis, Insect repellents: Concepts of their mode of action relative to potential sensory mechanisms 
in mosquitoes (Diptera: Culicidae), Journal of Medical Entomology, 22, 237, 1985. 

206. R. C. Smallegange, Y. T. Qiu, J. J. van Loon, and W. Takken. Synergism between ammonia, lactic acid 
and carboxylic acids as kairomones in the host-seeking behavior of the malaria mosquito Anopheles 
gambiae sensu stricto (Diptera: Culicidae), Chemical Senses, 30, 145, 2005. 

207. Y. T. Qiu. R. C. Smallegange. S. Hoppe, J. J. van Loon, E. J. Bakker, and W. Takken, Behavioral and 
electrophysiological responses of the malaria mosquito Anopheles gambiae Giles sensu stricto (Diptera: 
Culicidae) to human skin emanations. Medical and Veterinary Entomology, 18, 429, 2004. 

208. W. R. Mukabana, W. Takken, G. F. Killeen, and B. G. Knols, Allomonal effect of breath contributes to 
differential attractiveness of humans to the African malaria vector Anopheles gambiae. Malaria 
Journal, 3, 1, 2004. 

209. W. R. Mukabana, W. Takken, R. Coe, and B. G. Knols, Host-specific cues cause differential 
attractiveness of Kenyan men to the African malaria vector Anopheles gambiae. Malaria Journal, 1, 
17, 2002. 


2006 by Taylor & Francis Group, LLC 



28 


Insect Repellents: Principles, Methods, and Uses 


210. U. R. Bernier, D. L. Kline, C. E. Schreck, R. A. Yost, and D. R. Barnard, Chemical analysis of human 
skin emanations: Comparison of volatiles from humans that differ in attraction of Aedes aegypti 
(Diptera: Culicidae), Journal of the American Mosquito Control Association, 18, 186, 2002. 

211. T. Dekker, B. Steib, R. T. Carde, and M. Geier, L-Lactic acid: A human-signifying host cue for the 
anthropophilic mosquito Anopheles gambiae. Medical and Veterinary Entomology, 16, 91, 2002. 

212. T. Dekker, W. Takken, and M. A. Braks, Innate preference for host-odor blends modulates degree of 
anthropophagy of Anopheles gambiae sensu lato (Diptera: Culicidae), Journal of Medical Entomology, 
38, 868, 2001. 

213. H. V. Pates, W. Takken, K. Stuke, and C. F. Curtis, Differential behavior of Anopheles gambiae sensu 
stricto (Diptera: Culicidae) to human and cow odors in the laboratory. Bulletin of Entomological 
Research, 91, 289, 2001. 

214. E. Bare, I. N. Dogan, and P. A. Rossignol, An olfactometer for discriminating between attraction, 
inhibition, and repellency in mosquitoes (Diptera: Culicidae), Journal of Medical Entomology, 36,788, 
1999. 

215. A. Cork and K. C. Park, Identification of electrophysiologically active compounds for the malaria 
mosquito Anopheles gambiae, in human sweat extracts, Medical and Veterinary Entomology, 10, 269, 
1996. 

216. O. J. Bosch, M. Geier, and J. Boeckh, Contribution of fatty acids to olfactory host finding of female Aedes 
aegypti. Chemical Senses, 25, 323, 2000. 

217. L. C. Rutledge, M. A. Moussa, and C. J. Belletti, An in vitro blood-feeding system for quantitative 
testing of mosquito repellents. Mosquito News, 36, 283, 1976. 

218. L. C. Rutledge, R. A. Wirtz, M. D. Buescher, andZ. A. Mehr, Mathematical models of the effectiveness 
and persistence of mosquito repellents, Journal of the American Mosquito Control Association, 1, 56, 
1985. 

219. FIFRA, Insect repellent product performance testing guideline evaluation SAP report no. 02-02B, in 
Federal insecticide, fungicide and rodenticide act scientific advisoiy panel meeting, M. A. Thrall (Ed.), 
Arlington, VA: Environmental Protection Agency, p. 46, 2002, http://www.epa.gov/scipoly/sap/2000/ 
april/freportapril572000.pdf. 

220 A. Cockcroft, J. B. Cosgrove, and R. J. Wood, Comparative repellency of commercial formulations of 
DEET, permethrin and citronellal against the mosquito Aedes aegypti, using a collagen membrane 
technique compared with human arm tests, Medical and Veterinary Entomology, 12, 289, 1998. 

221. L. C. Rutledge, M. A. Moussa, C. A. Lowe, and R. K. Sofield, Comparative sensitivity of mosquito 
species and strains to the repellent diethyl toluamide, Journal of Medical Entomology, 14, 536, 1978. 

222. S. J. Moore, C. R. Davies, and M. M. Cameron, Research in Progress: Quantifying diversion of malaria- 
vector mosquitoes from repellent treated ‘bait' to unprotected ‘collectors’ in field tests in the Bolivian 
Amazon. Transactions of the Royal Society of Tropical Medicine and Hygiene, 97(6), 629, 2003. 

223. S. H. Nguyen, T. V. Nguyen. D. Phan Chau, and R. P. Marchand, Repellent cream with DEET: An 
effective and cheap method to reduce bites of An. dims species A in the forest, First ASEAN Congress of 
Parasitology and Tropical Medicine, Kuala Lumpur: Malaysian Society of Parasitology and Tropical 
Medicine, 2004. 

224. M. T. Gillies and T. J. Wilkes, Range of attraction of single baits for some West-African mosquitoes. 
Bulletin of Entomological Research, 60, 225, 1970. 

225. J. A. Hill, P. B. Robinson. D. L. Mcvey, W. A. Akers, and W. G. Reifenrath, Evaluation of mosquito 
repellents on the hairless dog. Mosquito News, 39, 307, 1979. 

226. S. Kasman, L. A. O. Roadhouse, and G. F. Wright, Studies in testing insect repellents, Mosquito News, 
13, 116, 1953. 

227. K. Watanabe, Y. Takada, N. Matsuo, and H. Nishimura, Rotundial, a new natural mosquito repellent 
from the leaves of Vitex rotundifolia. Bioscience, Biotechnology and Biochemistry, 59, 1995. 

228. N. Nicolaides, H. C. Fu, and G. R. Rice, The skin surface lipids of man compared with those of eighteen 
species of animals, Journal of Investigative Dermatology, 51, 83, 1968. 

229. Z. X. Li, J. A. Pickett, L. M. Field, and J. J. Zhou, Identification and expression of odorant-binding 
proteins of the malaria-carrying mosquitoes Anopheles gambiae and Anopheles arabiensis. Archives of 
Insect Biochemistry and Physiology, 58, 175, 2005. 


© 2006 by Taylor & Francis Group, LLC 


History of Insect Repellents 


29 


230. L. C. Rutledge, R. K. Gupta, R. A. Wirtz, and M. D. Buescher, Evaluation of the laboratory mouse 
model for screening topical mosquito repellents. Journal of the American Mosquito Control 
Association, 10, 565, 1994. 

231. M. G. Novak, W. G. Berry, and W. A. Rowley, Comparison of four membranes for artificially 
bloodfeeding mosquitoes. Journal of the American Mosquito Control Association, 7, 327, 1991. 

232. L. C. Rutledge, N. M. Echano, and R. K. Gupta, Responses of male and female mosquitoes to repellents 
in the World Health Organization insecticide irritability test system, Journal of the American Mosquito 
Control Association, 15, 60, 1999. 

233. D. R. Barnard, Global Collaboration for Development of Pesticides for Public Health: Repellents and 
Toxicants for Personal Protection, Geneva: World Health Organization, 2000. 

234. World Health Organization, Report of the WHO informal consultation on the evaluation and testing of 
insecticides, Geneva: World Health Organization, 1996. 

235. U.S. Environmental Protection Agency, Product performance test guidelines OPPTS 810.3700 insect 
repellents for human skin and outdoor premises “Public Draff", Washington, DC: United States 
Environmental Protection Agency Prevention, Pesticides and Toxic Substances, http://www.epa.gov/ 
opptsfrs/publications/OPPTS_Harmonized/810_Product_Performance_Test_Guidelines/Drafts/ 
810, 1999. 

236. R. K. Gupta and L. C. Rutledge, Laboratory evaluation of controlled-release repellent formulations on 
human volunteers under three climatic regimens, Journal of the American Mosquito Control 
Association, 5, 52, 1989. 

237. R. D. Xue, D. R. Barnard, and C. E. Schreck, Influence of body size and age of Aedes albopictus on 
human host attack rates and the repellency of DEET, Journal of the American Mosquito Control 
Association, 11, 50, 1995. 

238. R. D. Xue and D. R. Barnard, Human host avidity in Aedes albopictus : Influence of mosquito body size, 
age, parity, and time of day. Journal of the American Mosquito Control Association, 12, 53, 1996. 

239. D. R. Barnard, K. H. Posey, D. Smith, and C. E. Schreck, Mosquito density, biting rate and cage size 
effects on repellent tests, Medical and Veterinary Entomology, 12, 39, 1998. 

240. D. R. Barnard, Mediation of DEET repellency in mosquitoes (Diptera, Culicidae) by species, age and 
parity, Journal of Medical Entomology, 35, 340, 1998. 

241. R. D. Xue andD. R. Barnard, Effects of partial blood engorgement and pretest carbohydrate availability 
on the repellency of DEET to Aedes albopictus. Journal of Vector Ecology, 24, 111, 1999. 

242. I. A. Klun and M. Debboun, A new module for quantitative evaluation of repellent efficacy using human 
subjects, Journal of Medical Entomology, 37, 177, 2000. 

243. R. Natarajan, S. C. Basak, A. T. Balaban, I. A. Klun, and W. F. Schmidt, Chirality index, molecular 
overlay and biological activity of diastereoisomeric mosquito repellents, Pest Management Science, 61, 
1193, 2005. 


© 2006 by Taylor & Francis Group, LLC 


© 2006 by Taylor & Francis Group, LLC 



2 


Terminology of Insect Repellents 


Graham B. White 


CONTENTS 

Basic Repellent Terminology.31 

Glossary and Definitions for Repellent Science.32 

References.43 


Smell is fatal for repellents intended to be used in jungle warfare but, provided it is pleasant, it may 
even be an advantage in civilian use. Owing to the importance attached to long duration of 
effectiveness for military purposes, research on repellents during the war has tended to develop a 
type of repellent with very high boiling-point and hence, almost as a corollary, less effective at 
a distance than some more volatile repellents. 

(Christophers, 1947)' 


Basic Repellent Terminology 2 4 

The English word repellent is a noun (the repellent material) or an adjective (repellent effect), derived 
from the Latin verb repellere , meaning “to drive back," the movement away being repulsion. The 
alternative spelling repellant, with an “a,” comes from -antem meaning “an agent of action.” 
Attractant has the opposite meaning, based on the Latin attractum for being pulled towards 
something. The word attractant is a noun (something that attracts) or an adjective (being attractive), 
depending on the context and syntax, etymologically derived from the Latin verb trahere, meaning 
“to draw or pull.” Therefore, anything that attracts or repels particular insects is either an insect 
attractant or an insect repellent. Generally, for chemicals affecting feeding behavior negatively or 
positively, by any mode of action, this book introduces the new term phagomone, as discussed in the 
Introduction and the Epilogue. Some materials and physical factors (e.g., heat and light) can elicit 
either repellent or attractant effects, depending on quantitative factors (Chapter 9) and circumstances. 

To help foster scientific perceptions, Dethier 5-7 defined repellents as “any stimulus which elicits an 
avoiding reaction” and made a further distinction, in terms of the physical state of the chemical, by 
recognizing contact repellents and vapor repellents, i.e., those that have to be touched by the insect or 
simply detected in the air. Differentiating these modes of exposure remains challenging, as discussed in 
Part 2 of this book, because the treatment distinction may not be absolute. Generally, to achieve personal 
protection with some duration of effectiveness, repellents are applied ad libitum to chosen parts of the 
skin and clothes; due to this topical treatment (derived from the Greek word topos, meaning “limited 


31 


© 2006 by Taylor & Francis Group, LLC 










32 


Insect Repellents: Principles, Methods, and Uses 


location”) they are commonly known as topical repellents. Some devices, e.g., mosquito coils and 
repellent vaporizers, are designed to protect an area outdoors or volume of space indoors by releasing 
spatial repellent 8,9 vapor for as long as the device operates (Chapter 23), but their effectiveness quickly 
fades when emission stops and the repellent dissipates. 

Commercially, insect repellents are consumer products marketed in every society through suitable 
retailers (e.g., camping and travel shops, pharmacies, supermarkets) and by mail order. The traditional 
repellent business became more scientifically rigorous when synthetic chemicals began to replace 
botanicals as the products-of-choice during the 1940s and 1950s. Previously, so-called “culicifuges” 
and repellents to ward off noxious arthropods comprised a wide variety of popular natural products 
(Chapter 14 and Chapter 15), few of which had been evaluated entomologically or standardized for 
efficacy. The repellent market grew and evolved rapidly following the 1939-1945 World War II 
period, thanks to results of intense research efforts to discover and develop repellents for military use, 
as described in Chapter 1. Hence the technical foundations of repellent science were mainly 
established by three loosely coordinated groups: working in Rutgers, 10 ' 11 New Jersey, USA, 
Cambridge, 12 ' 13 UK, and Orlando, 14-17 Florida, USA, continuing to this day at Gainesville, 18 
Florida, USA. They developed standardized testing methods with mosquitoes (Aedes aegypti ) and 
ticks ( Amblyomma americanum ) that still provide the basis of screening procedures and comparative 
assessment of repellents (Chapters 5-9). 

The following glossary attempts to explain the meanings of a wide range of terms needed to 
understand repellent science and associated research. This list and supporting references augment the 
greater attention given to the major topics in successive chapters of this book. Included here are the 
acronyms for relevant organizations and regulatory statutes. The Index provides further reference to key 
words and Appendix 2 provides details on the appropriate chemical designations for many of the 
active ingredients. 


Glossary and Definitions for Repellent Science 

abiotic factors Pertaining to repellents: non-biological variables that may influence repellency, 
e.g., air quality, humidity, light, temperature, wind (discussed in Chapters 5, 8, and 12); 

c.f. biotic factors. 

absorb; absorption The process by which repellent enters a substrate, e.g., skin (c.f. adsorb). 

acidity pH < 7. 

activator Something (e.g., heat, synergist, volatile solvent) that, when added to or combined with a 
repellent, increase its availability or activity (c.f. synergist). 

active ingredient (a.i.); active material See below, under ingredient. 

adjuvant Inert chemical added to repellent formulation to enhance its effectiveness. 

adsorb; adsorption The process by which repellent is bound to the surface of a particle or 
absorbent substance. 

aerosol Extremely fine spray droplets suspended in air. The WHO 19 classifies spray droplets as fine 
aerosols < 25 pm, coarse aerosols 25-50 pm, mists 50-100 pm, fine sprays 100-200 pm, 
medium sprays 200-300 pm and coarse sprays > 300 pm. 

aggregate To gather together, assemble. 

alkalinity pH > 7. 

allelochemicals Non-nutritional semiochemicals used by one species to affect (behavior, feeding, 
growth, health, breeding of) another species. 

allomone Chemical substance (produced or acquired by an organism) that, when contacting an 
individual of another species, evokes in the receiver a behavioral or physiological reaction 
adaptively favorable to the emitter (opposite of kairomone). 

antagonism; antagonist Reduction of the potency of a repellent; that which causes antagonism. 


© 2006 by Taylor & Francis Group, LLC 




Terminology of Insect Repellents 


33 


anthropophagous; anthropophagy Feeding on humans (c.f. Zoophagy). 

anthropophilic; anthropophily Tendency of hematophagous anthropods to prefer human hosts. 
AOAC Association of Official Analytical Chemists International (http://www.aoac.org), founded 
1884, oversees the most extensive program for validation of Official Methods of Analysis 
(OMAs), but none specifically for repellents (c.f. CIPAC). 
aqueous Dilution in water. 

arrestant Chemical that causes insects to aggregate in contact with it, the mechanism of 
aggregation being kine (by movement) or having a kinetic component. 7 An arrestant may 
slow the linear progression of the insects by reducing actual speed of locomotion or by 
increasing turning rate (c.f. locomotor stimulus). The -ant form of this word is 
etymologically correct (not arrestent) because arrest is derived, through Old French, from 
the vulgar Latin arrestare. 

arthropods Invertebrate Phylum Arthropoda. Creatures with exoskeleton (consisting of chitin) and 
jointed legs. The blood-feeding arthropods are either insects (Class Insecta) or mites/ticks 
(Class Arachnida, Order Acari). Numerous other groups of animals affect humans directly 
through bites or envenomation (e.g., snakes, scorpions, spiders, and wasps), 
attractant For insects, something that causes (attraction) insects to make oriented movements 
towards its source 7 —i.e., the opposite of repellent (Chapter 9). Associated terms: (verb) to 
attract; (nouns) attractance, the quality of attracting; attraction, the act of attracting or the 
state of being attracted; (adjective) attractive, serving to attract. Sex attractant, substance or 
mixture of substances released by an organism to attract members of the opposite sex of the 
same species for mating. 

behavioristic avoidance 20 Also known as behavioristic resistance or protective avoidance— 
modified behavior whereby endophilic mosquito populations sometimes adapt to exophily 
in response to pressure of indoor residual spraying with excitorepellent insecticide, 
bioassays Standard methods and procedures for replicated comparative testing of effects on 
biological materials. 21-23 Chapter 6 and Chapter 9 describe bioassays for attractants 
and repellents. 

Biocidal Products Directive of the European Commission Regulatory law for pesticides in all 
countries of the European Union, implemented by national governments and the E.U. 
Environment Directorate (http://europa.eu.int/comm/environment/biocides/index.htm). 
This Directive 98/8/EC of the European Parliament and of the Council on the placing on 
the market of biocidal products was adopted in 1998. According to the directive, member 
states had to transpose the rules before 14 May 2000 into national law. The Biocidal Product 
Directive aims to harmonize the European market for biocidal products and their active 
substances. At the same time, it aims to provide a high level of protection for humans, 
animals, and the environment. The Commission adopted the original proposal for the 
Biocidal Products Directive in 1993, following the model established by Directive 91/414/ 
EEC on plant protection products, adopted in 1991. 
biotic factors Pertaining to repellents. Biological variables that may influence repellency, such as 
physiological condition of the insect (e.g., level of hunger, activity cycle) or the host (e.g., 
rates of exhalation and sweating), as discussed in Chapter 5; c.f. abiotic factors, 
biting rate The number of bites/person/time period (e.g., 12 bites/hour), as a measure of population 
density in relation to humans, for any given species of biting arthropods, or group of species 
at a particular place and time. For ethical reasons, especially where vector-borne disease 
risks must be considered, it is customary to intercept the attacking insects before they 
actually bite (possibly increasing catch efficiency); the results are therefore reported in terms 
of the “landing rate” rather than the biting rate. The coefficient of protection 24 (CP), is given 
by [(A — B)/A\X 100, where A is the average number biting the untreated person per hour 
and B is the average number biting the experimentally treated subject during the same 
exposure period and conditions; CP is commonly used to assess the relative effectiveness of 


© 2006 by Taylor & Francis Group, LLC 


34 


Insect Repellents: Principles, Methods, and Uses 


candidate materials compared to deet. Other criteria for repellent testing under field 
conditions are the period of time to first bite, or first confirmed bite, or duration of a 
reduction in biting—the choice of criterion depending inter alia on the local biting rate 
pressure. 25 Considerable disagreement exists on the appropriate measurement of repellent 
product efficacy, as discussed throughout this volume, 
botanical Pertaining to green plants (Embryophytes): plant sources of repellent natural products, 
butyl carbitol acetate Also known as diethelene glycol monobutyl ether acetate. This compound 
was the standard of comparison adopted by Granett 11 ' 12 (1940) for screening repellents at 
the Orlando Institute, Florida, precursor of the Insects Affecting Man and Animals Research 
Laboratory at Gainesville, Florida, now the Center for Medical, Agricultural and Veterinary 
Entomology, of the Agricultural Research Service of the U.S. Department of Agriculture, 
carrier Inert solid or liquid material used to prepare repellent formulation. 

CAS numbers Unique numerical identifiers for chemical compounds, polymers, mixtures and 
biological sequences. Chemical Abstracts Service (CAS), a division of the American 
Chemical Society, assigns these identifiers to every chemical described in the literature. 
They are also called CAS registry numbers (CAS RNs). Substances also receive unique CA 
index names, constructed using rigid nomenclature rules. In an effort to facilitate searching 
for related compounds, the most important functional groups of a substance are named first, 
followed by their modifications (c.f. IUPAC names). http://www.cas.org/EO/regsys.html. 
CDC Centers for Disease Control and Prevention, U.S. Department of Health and Human Services. 

CDC policy and guidelines 26,27 for repellents are issued by the Division of Vector- 
Borne Infectious Diseases, and implemented by the National Center for Infectious Diseases 
(based at Fort Collins, Colorado, USA), and by the Entomology Branch (based at Atlanta, 
Georgia, USA). 

CFR Code of Federal Regulations of the United States of America (USA): http://www.gpoaccess. 

gov/cfr/index.html. Concering pesticides, including repellents, Title 21 deals with FDA, 
including GRAS materials; Title 40 deals with EPA including FIFRA and FQPA 
(Chapter 26). 

CIPAC Collaborative International Pesticides Analytical Council (http://www.cipac.org). The reco¬ 
gnized international, nonprofit, and non-governmental organization, promotes international 
agreement on methods for the analysis of pesticides and physico-chemical test methods for 
formulations. Methods are proposed by manufacturers (Companies) and are tested 
internationally by the inter-laboratory program for evaluation of test methods. After 
validation of analytical results and adoption, the methods are published in CIPAC 
Handbooks. 

compatible Ingredients that retain their individual properties when mixed together, 
concentrate Chemical formulation containing a high percentage of active ingredient (a.i.). 
concentration Proportion of a given ingredient in a formulation or solution, e.g., oz/gal, mg/L. 
cosmetic ( adj .): Serving to beautify, or («.): a preparation for beautifying the face, hair, skin, etc. 

Chapter VI of the U.S. Federal Food, Drug, and Cosmetic Act (FD&C Act of 1906, Title 21 
of the U.S. Code, plus amendments, currently administered by the FDA) defines cosmetics 
as articles intended to be applied to the human body for cleansing, beautifying, promoting 
attractiveness, or altering the appearance without affecting the body’s structure or functions. 
Included in this definition are products such as skin creams, lotions, perfumes, lipsticks, 
fingernail polishes, eye and facial make-up preparations, shampoos, permanent waves, hair 
colors, toothpastes, deodorants, and any material intended for use as a component of a 
cosmetic product. Soap products consisting primarily of an alkali salt of fatty acid and 
making no label claim other than cleansing of the human body are not considered cosmetics 
under U.S. law. Likewise, insect repellents are not cosmetic products, although it would be 
possible to include repellent active ingredients in particular cosmetics, as done with some 
“sun screen” anti-UV preparations combined with deet (that enhances absorption, raising 


© 2006 by Taylor & Francis Group, LLC 


Terminology of Insect Repellents 


35 


systemic toxicity) 28 marketed for giving skin protection against both sunburn and biting 
insects. The term cosmetic properties of a repellent product is often used to describe the 
properties of the formulation that do not affect performance, but that alter the subjective 
perception of the product (e.g., fragrance, oiliness, color). 

CSPA Consumer Specialty Products Association represents the interests of the consumer specialty 
products industry in the U.S., providing households, institutions and industrial customers 
with products for a cleaner and healthier environment (http://www.cspa.org). 

CTFA Cosmetic, Toiletry, and Fragrance Association (http://www.ctfa.org), publisher of the 
International Cosmetic Ingredient Dictionary and Handbook, 29 giving International Nomen¬ 
clature Cosmetic Ingredient (INCI) names for cosmetics and personal care products, e.g., 
EBAAP for IR3535. 

culicifuge 30,31 Repellent for use against mosquitoes (Culicidae), the suffix based on the Latin verb 
fugere, meaning “to flee.” 

deet /V,/V-diethyl-3-methylbenzamide (originally known as /V,/V-diethyl-/wefa-toluamide), usually 
abbreviated to deet or deet in literature. It is the dominant repellent used worldwide since the 
1960s. Globally it is the leading active ingredient of insect repellent products, being 
effective against all groups of biting arthropods and even leeches. Formulations containing 
from 4% to 100% deet are registered by the EPA for direct skin application to repel insects, 
rather than kill them. Deet is registered for use by consumers, plus a few veterinary uses, but 
is not used on food. Market surveys in the U.S. show that about a third of the population use 
deet-based products, currently available to the public in a variety of liquids, lotions, sprays, 
and impregnated materials (e.g., wipes and wrist bands). After it was discovered by the 
USDA Agricultural Research Service and developed by the U.S. Army in 1946, deet was 
introduced for use by the general public in 1957. More than 230 products containing deet 
(CAS# 134-62-3) are currently registered with EPA by more than 70 companies (http://deet. 
com and http://www.deetonline.org). Further details on deet are given in Chapter 16. 

deterrent (n. or cidj.) In the repellent context, something that inhibits feeding or oviposition when 
present in a place where insects would, in its absence, feed or oviposit. 7 In the biological 
context, something that protects against bodily harm: see Chapter 3 and Berenbaum (1995) 
for deterrent chemicals. 32 Associated terms include deter (v.): to discourage or prevent, and 
deterrence ( n .): the act of deterring. These terms fit the way that permethrin-impregnated 
materials (e.g., clothes or bednets) deter blood-thirsty female mosquitoes, etc. from biting, 
or even from entering a house 33 ; whereas, other pyrethroid treatments are more insecticidal 
than deterrent or repellent (c.f. excitorepellent). 

diluent Material used to reduce concentration of an active ingredient in a formulation, e.g., 
dilution of concentrate to make the operational concentration. 

DMP Dimethyl phthalate (CAS# 131-11-3), an insect repellent with many other uses as a plasticizer 
and in solid rocket propellants. Commercially, DMP was superseded by deet, DEPA, PMD, 
and others (chapters 1 and 22) for repellent markets. 

dispersing agent Material that reduces the attraction between particles. 

dosage Quantity of active ingredient applied per unit of time (e.g., 10 oz/day) or area (1 cm/m 2 ) or 
volume (e.g., 1 mg/L) or personal application (e.g., 1 mL/arm/day). See Chapters 6, 8, 12 for 
dosage criteria employed for comparative evaluation of repellents, including the effective 
dose (actual concentration) giving 50% or 90% reduction of biting (ED 50 and ED 90 ) and the 
minimum effective dose to prevent biting completely, these bioassay parameters are mostly 
employed for comparative studies in the laboratory; see biting rate for field criteria, 
discussed in Chapters 6, 8 and 12. 


Table 25.1 includes one such product marketed in the U.S., whereas the Canadian Pest Management Regulatory Agency 
ruled (RRD2002-01) against their acceptability for registration, due to incompatible rates of application (i.e., deet should be 
applied sparingly, whereas sunscreens should be applied liberally: www.pmra-arla.gc.ca/english/pdf/iTd/rrd2002-01-e.pdf). 


© 2006 by Taylor & Francis Group, LLC 



36 


Insect Repellents: Principles, Methods, and Uses 


EBAAP Ethyl butyl acetyl aminopropionate (INCI name); chemical description 3-(/V-«-butyl-/V- 
acetyl)-aminopropionic acid ethyl ester; derived synthetically from (3-alanine (a natural 
amino acid); commercially known as IR3535®. Approved by the WHOPES 34 and interim 
specifications issued. 

ECB European Chemicals Bureau, responsible inter alia for the Biocidal Products Directive (q.v.) 
of the European Commission (Chapter 26 and http://ecb.jrc.it/biocides/). 

EDTIAR Extended Duration Topical Insect and Arthropod Repellent (deet-based slow-release 
formulation) introduced in 1990 for U.S. military use; commercially marketed as 
Ultrathon™ 4 (http://www.ultrathon.com). 

emulsifier A chemical that aids in the suspension of one liquid in another. 

endophagic; endophagy Feeding indoors by endophilic mosquitoes etc. 

endophilic; endophily Tendency of insects (especially female Anopheles mosquitoes of some 
species) to come into houses for biting nocturnally and resting diurnally (opposite of 

exophily). 

entomology The study of insects; commonly assumed to include other arthropods (q.v.). 

EPA U.S. Environmental Protection Agency, see USEPA below and Chapter 26. 

essential oils Terpenes and other volatiles obtained from plants by steam distillation or pressing, 
they are hydrophobic and mostly aromatic. Many are repellent to insects and some are 
potent insecticides; traditionally they have been employed as pesticides around the world. 35 
Encouraged by the EPA 1996 exemption to FIFRA for minimum risk pesticides, many have 
recently been developed and commercialized as pesticides in the USA. Among the most 
effective 36 as repellents are white cedar oil (CAS# 8000-34-8), peppermint oil (CAS# 806- 
90-4), red thyme oil (CAS# 8007-46-3), bourbon geranium oil (CAS# 8000-46-2), linalool 
(Appendix 2), and dehydrolinalool. 37 However, as indicated in Chapter 14 (pp. 292-293), 
toxicological risk assessment is necessary to establish safety and tolerance levels for 
essential oils used as repellents or in foodstuffs [FDA category (CFR 21: 170) Generally 
Regarded as Safe (GRAS)]. 

EU The European Union (EU) of twenty-five countries (2006) with 20 official languages, formerly 
known as the European Community (EC), originally the European Economic Community 
(EEC). The Biocidal Products Directive (q.v.) determines pesticide regulatory status 
throughout the EU (http://europa.eu.int/). 

evaporate (v.) To change from solid or liquid to vapor (U.S.) or vapour (U.K.), synonymous with 
vaporize (U.S.) or vapourise (U.K.); evaporation («.): The process of evaporating; 
evaporate to dryness. 

excitorepellency The power of DDT and some pyrethroids, especially through tarsal exposure of 
insects, to irritate them sufficiently that they fly away before knockdown, even from 
sublethal exposure; 20 ' 38-40 thereby adult female mosquitoes become more exophilic instead 
of endophilic and this contributes to greater reduction of their vectorial capacity than from 
simply killing a lesser proportion of the vector population. 41 

exophagous; exophagy Behavioral tendency of female mosquitoes etc. to bite hosts outdoors. 

exophilic; exophily Tendency of most insects to stay outside buildings (contrasts with endophily 
for malaria vector Anopheles females that enter houses to bite and take shelter). 

FDA Food and Drug Administration of the U.S. Department of Health and Human Services, having 
regulatory responsibility for cosmetics and medicines etc., but not for insect repellents 
(http://www.fda.com/). 

FIFRA The U.S. Federal Insecticide, Fungicide, and Rodenticide Act (1947, 1972 and amend¬ 
ments) for pesticides regulation (40 CFR), administered by the EPA. 


A registered trademark of Merck KGaA, Darmstadt, Germany. 
1 A registered trademark of 3M Corporation, St. Paul. MN. 


© 2006 by Taylor & Francis Group, LLC 



Terminology of Insect Repellents 


37 


formulation in.) Defined chemical product mixture, usually meaning the commercialized version 
of a special formula, sometimes requiring dilution before use. 

FQPA Food Quality Protection Act (U.S. Public Law 104-170, 1996: http://www.epa.gov/ 
oppfeadl/fqpa/backgrnd.htm), augmenting FFDCA and FIFRA, administered by the 
USEPA fq.v.): intensifies regulatory controls on pesticides for reasons of human and 
environmental health (Chapter 26 summarizes EPA role under Title 40, parts 150-189, of 
the U.S. code of Federal Regulations). 

GFP and GLP Good Field Practice and Good Laboratory Practice, internationally recognized 
standards of conduct and procedure, administered by the Organisation for Economic 
Co-operation and Development (OECD), to ensure the generation of high quality and 
reliable test data related to the safety of industrial chemical substances and preparations in 
the framework of harmonizing testing procedures for the Mutual Acceptance of Data 
(MAD) (http://www.oecd.org/document/63/0,2340,en_2649_34381_2346175_l_l_l_l,00. 
html). 

Generally Regarded as Safe, classification by FDA, www.cfsan.fda.gov/~lrd/cfrl7030.html 
<http://www.cfsan.fda.gov/~lrd/cfrl7030.html>, similar to minimum risk classification 
by EPA (Chapter 26, p. 420). 

Potential source of harm. For repellents and other pesticides, the World Health Organiz¬ 
ation (WHO) classification 68 based on the rat LD 50 by weight, following oral or dermal 
exposure, assuming solids are four-fold more hazardous than liquids, recognizes the 
following categories: class la, extremely hazardous; class lb, highly hazardous; class II, 
moderately hazardous (e.g., DDT, permethrin, pyrethrins); class III, slightly hazardous 
(e.g., deet); plus active ingredients unlikely to cause acute hazard in normal use. 
hematophagous arthropods Blood-feeding insects, ticks and mites. English spelling: haemato- 
phagous. Commonly referred to as “biting insects.” 
hydrogen-ion concentration Usually expressed as the negative log (pH), a measure of acidity- 
alkalinity. 

immiscible Liquids that cannot mix to form homogeneous solution. 

INCI See CTFA. 

incompatible Ingredients that do not retain their individual properties when mixed together, 
ingredient That which goes into a compound, formulation, preparation, or mixture; active 
ingredient (a.i.), the key ingredient with intended activity, 
inhibition As discussed in Chapter 4, activity-inhibitors cause a neutral reaction, neither attraction 
nor repulsion, whereby an insect fails to proceed questing purposefully, but is not 
anaesthetized nor narcotized. Dogan and Rossignol describe an olfactometer 42 for 
discriminating between attraction, inhibition, and repellency in mosquitoes, 
insect Any member of the arthropod Class Insecta. The name derived from the Latin insectum for 
having been cut, referring to the articulated body; adults typically with three pairs of legs 
(hexapod). 

insecticide Chemical agent used to kill insects; mostly suitable for use also as acaricides. 
insoluble Inability of a substance to dissolve in a particular liquid solvent. 

irritancy The power of DDT and some pyrethroids (especially those with alpha-cyano moiety) to 
irritate arthropods, causing excitorepellency (q.v.). 

IR3535 See EBAAP: insect repellent. 

IUPAC International Union of Pure and Applied Chemistry (an international non-profit, non¬ 
governmental organization for the advancement of chemistry, consisiting of national 
chemistry societies (http://www.iupac.org). IUPAC is the recognized authority in develop¬ 
ing standards for naming the chemical elements and their compounds, through its Inter- 
Divisional Committee on Nomenclature and Symbols (IUPAC nomenclature), c.f. 
CAS, CIPAC. 


GRAS 

hazard 


© 2006 by Taylor & Francis Group, LLC 


38 


Insect Repellents: Principles, Methods, and Uses 


kairomone A substance released by one species that benefits members of another (e.g., parasites 
detect host kairomones), by being a signal or attractant to them (opposite of allomone), 
kinesis non-oriented movement of an organism; c.f. taxis (oriented movement); kinetic (adj.). 
knockdown Sublethal incapacitation; early symptom of an insect responding to a pesticide; not 
necessarily lethal because metabolic recovery may occur. Hence the rates of knockdown and 
mortality are scored separately, usually 1 hour and 24 hours post-treatment in bioassays. 
Knockdown has another meaning in molecular biology, for gene incapacitation, 
locomotor stimulant A chemical that causes, by a kinetic mechanism, insects to disperse from a 
region more rapidly than if the area did not contain the chemical. The effect may be to 
increase the speed of locomotion, to cause the insects to carry out avoiding reactions, or 
to decrease the rate of turning. 43 

market value Globally and locally, the price of repellent products is determined by market forces, 
whereas the sale cost (over-the-counter price) of each repellent unit (pack) includes the 
values of active ingredients, formulation ingredients, manufacturing and labor, packaging, 
distribution, promotion, sales and profit margins, plus taxes and tariffs. World-wide the 
global market value 70 of repellents was estimated at $2 billion in 2002. 
mortality rate Proportion of sample killed in a test (usually scored 24 hours after treatment) by 
exposure to a lethal dose causing fatality; those surviving treatment have experienced only a 
sub-lethal dose, that may affect their bionomics and behaviour, e.g., inhibition, deterrence, 
and repulsion. 

natural products Exploitable materials formed by nature, including foodstuffs and natural fibres 
used for weaving fabric, e.g. cotton. Natural repellent products from plants (botanically- 
derived) are reviewed in Chapters 14 and 15: those from non-woody plants are herbal-based 
(Chapter 9). Natural pyrethrins comprise important insecticides and repellents, 
organic Strictly, chemical compounds derived from plants or animals, plus other carbon-based 
materials. Essential oils from plants (Chapter 14) include many useful organic repellents. In 
the terminology of modern farming and horticulture, so-called “organic” vegetables and 
other agricultural produce are defined as those grown and marketed without application of 
synthetic pesticides. 

personal protective measures 44 (PPM) Protective measures against biting arthropods, such as the 
personal use of repellents, bednets and clothing. 

pesticides Chemicals for killing pests, classified by EPA as follows: algicides, antifouling agents, 
antimicrobials, biocides, biopesticides, defoliants, desiccants, disinfectants and sanitizers, 
fungicides, fumigants, herbicides, insect growth regulators, insecticides, acaricides 
(including miticides), molluscicides, nematicides, ovicides, pheromones, plant growth 
regulators, rodenticides, and repellents (http://www.epa.gov/pesticides/about/types.htm). 
pH Number expressing degrees of acidity (pH<7) and alkalinity (pH >7) in solutions; pH 7 is 
neutral. Mathematically, pH is the log 10 of the reciprocal of the hydrogen ion concentration; 
it is usually measured by comparison with a standard solution of potassium hydrogen 
phthalate, with a pH of 4 at 15°C. 

phagomone Chemical that affects feeding behavior, negatively or positively, by any mode of action 
(see Preface and Epilogue). 

pheromone A chemical compound, emitted by an organism, that influences the behavior and 
development of other members of the same species, 
phytotoxicity Pathological effect on plant (Greek: phytos) vegetation. 

picaridin (KBR 3023) Insect repellent developed and commercialized as Bayrepel®; approved by 
WHOPES 34 and interim specifications issued under the proposed ISO name icaridin [sic], 
piperamides and piperidine alkaloids A series of compounds and analogs that includes many 
useful repellents, some being also insecticidal, e.g., deet, SS220 and pipernonaline. 45-48 The 
amides have a carbonyl (C=0) group linked to a nitrogen, N-(C=0), while 
the nitrogen’s other two bonds are linked with hydrogens (Figure 2.1) or other groups, 


© 2006 by Taylor & Francis Group, LLC 


Terminology of Insect Repellents 


39 



A/,/V-diethyl-3-methylbenzamide t -(3-methylbenzoyl)piperidine 

FIGURE 2.1 Deet (on the left) has a benzene ring linked by a carbonyl group (C=0) to an amide (piperamide) with two 
CH 3 methyl groups; the piperidine analog (on the right) has a saturated carbon ring that includes the nitrogen from the amide. 


e.g., (V,(V-diethyl-3-methylbenzamide (deet). When the nitrogen joins a saturated hetero¬ 
cyclic ring with five carbons, the compound constitutes a piperidine—the chemical name 
derived from plants of the pepper family (Piperaceae) that contain many such natural 
compounds, sometimes used as repellents (Chapter 14). Natural piperidine (CAS# 110-89- 
4) is the noxious ingredient of poison hemlock ( Conium maculatum ) in the carrot family 
Apiaceae. Among more than 200 such compounds identified in Piper, 49 the relatively simple 
amides provide much of the “hot pungent spice” taste as well as the biological activity in 
many species. 44 The piperamides commonly found in the genus Piper are bifunctional; an 
isobutyl amide functionality is combined with a methylenedioxyphenyl (MDP) moiety, as 
seen in piperine of Piper nigrum fruit and 4,5-dihydropiperlonguminine in foliage of the 
Central American Piper tuberculatum. The most active piperamide discovered to date is 
pipercide, approximately 100-fold more active than piperine. 50-52 The piperamides are also 
unusual because of their dual biological activities: the amide functionality is neurotoxic and 
the MDP group is an inhibitor of cytochrome P450 enzymes. Scott et al. 53 demonstrated that 
combinations of piperamides in binary, tertiary, and quarternary mixtures had successively 
higher toxicity at equimolar concentrations. This combination of useful traits suggests that 
Piper extracts may be good candidate pesticides with a rich range of insecticidal and 
repellent properties. 

PMD, p-menthane-3,8-diol Occurs naturally in leaves of the Australian lemon-scented gum tree 
(Corymbia citriodora), commonly called lemon eucalyptus. 54 This monoterpene, structu¬ 
rally similar to menthol (CAS# 42822-86-6), remains as a spent product after distillation of 
essential oils from leaves and twigs of Corymbia citriodora. Whereas natural PMD-based 
repellents have long been popular in China and elsewhere, 55 and registered in Europe for 
over a decade, synthetic PMD is used as the active ingredient for some of the repellents 
marketed as “lemon eucalyptus” in the U.S. As described in Chapter 20, PMD exerts 
repellency of the highest order against a wide range of hematophagous arthropods. 
Formulations registered in the USA include liquids that are sprayed on skin or clothing, 
or lotions that are rubbed on skin. Not yet submitted for WHOPES evaluation. 

PPM Acronym for Personal Protective Measures 44 against biting arthropods, such as the use of 
topical repellents and clothing (c.f. ppm expressing dilution in terms of parts per million). 

pyrethrins Oily esters extracted from cultivated pyrethrum flowers. Chrysanthemum cinerariae- 
folium Benth. & Hook., syn. Tanacetum cinerariifolium (Trevir); also found in pyrethrum 
daisies: Chrysanthemum roseum Web. & Mohr., syn. C. coccineum Willd. (Asteraceae). 
Crude pyrethrin extract contains three esters of chrysanthemic acid (chrysanthemates: 
pyrethrin I, cinerin I, jasmolin I) plus three esters of pyrethrin acid (pyrethrates: pyrethrin II, 
cinerin II, jasmolin II), combined ratio 71:21:7, generally known as pyrethrins. Being 
lipophilic but having low aqueous solubility, pyrethrins are readily absorbed via 
arthropod cuticle but not via the skin of vertebrates. Pyrethrins are very potent insecticidal 
knockdown agents, causing excitorepellency at sublethal doses, due to disruption of 


© 2006 by Taylor & Francis Group, LLC 


40 


Insect Repellents: Principles, Methods, and Uses 


sodium channel gating in myelinated nerves. Commercially, 25% to 50% pyrethrin 
concentrates are very stable in darkness at ambient temperatures, but degrade rapidly in 
sunlight (DT 50 10—12 mins). 

pyrethroids Numerous synthetic organic compounds, mostly based on the chrysanthemate moiety 
of pyrethrum, having analogous neurotoxic modes of action causing rapid knockdown and 
insecticidal effects. Discovery and development of synthetic pyrethroids, during 1960s and 
70s, accomplished several goals: more economical and consistent production than with 
natural pyrethrins; photostable products with residual efficacy but limited bioaccumulation. 
After early progress with allethrins (transient space sprays and vaporizers), the first truly 
stable pyrethroids were fen valerate and permethrin; their relative safety and potency greatly 
surpassed other classes of insecticides. Wide variations in potency occur between cis and 
trans isomers, and among enantiomers of pyrethroids, allowing much diversity of pyrethroid 
products, providing manufacturers and users with choices between knockdown versus 
insecticidal potency, and degrees of residual stability. As hundreds of pyrethroids became 
commercialized, this class of compounds has dominated the insecticide industry during 
recent decades. Permethrin remains one of the favorites for its versatility as an insecticide 
with repellent and deterrent properties (Chapters 5 and 6). Other pyrethroids mentioned in 
this book include allethrins, alpha-cypermethrin, beta-cyfluthrin deltamthrin, esbiothrin, 
lambda-cyhalothrin, metofluthrin, prallethin, tetramethrin and transfluthrin (Appendix 2). 

repellent, repellant For insects, something that causes insects to make oriented movements away 
from its source. 7 Associated terms: (verb) to repel; (nouns) repellency (repellancy), the 
quality of repelling; repeller, device for repelling (invalid for electronic 56 so-called 
“mosquito repellers”); repulsion, the act of repelling or the state of being repelled; 
(adjective) repulsive, serving to repel. The term repellent has received such general usage 
as a formulated product or as a chemical with a specific behavioral effect that it has lost 
much of its technical meaning. The editors of this volume advocate that the term repellent 
be restricted to the designation of products intended to reduce the rate of biting from 
hematophagous arthropods (French: insectifuges corporels). In this way, the technical 
literature will tend to use more precise terms that describe the effects of chemicals on 
specific behaviors. The introduction of the term phagomone is, in part, an attempt to 
facilitate this transition by providing the technical literature with an alternative to the term 
repellent used generally. 

resistance Defined by the WHO (1957) 57 as “the development of an ability in a strain of some 
organism to tolerate doses of a toxicant that would prove lethal to a majority of individuals 
in a normal (susceptible) population of the same species,” various types of insecticide 
resistance are well known in many species of flies, mosquitoes and other vectors and pests of 
public health importance. 58 For an increasing number of species, diagnostic and discrimi¬ 
nating dosages have been determined 59 ' 60 for distinguishing between susceptible and 
resistant individuals. Selection for resistance against repellents might be expected, due to 
their ubiquitous usage and environmental persistence 61 ' 62 of deet. Because no effort is made 
to monitor the sensitivity of wild populations of the many arthropod species that repellents 
are employed against, the possibilities of behavioral or physiological resistance to repellents 
remain unexplored. However, studies with laboratory strains of mosquitoes 63 ' 64 and 
Drosophila 65 demonstrate genetic selection of insensitivity and tolerance, indicating the 
potential for resistance to deet and other repellents. 

risk assessment In context of human health, estimating the probability of adverse effects resulting 
from defined exposure to known chemical hazard 68 - see (Chapter 14, pp. 292-293) and 
(Chapter 26, p. 422) for repellents. 

Rutgers 612 The original proprietary name for ethyl hexanediol (CAS# 94-96-2) when used as 
a repellent product; withdrawn 1991 for toxicological reasons (Chapter 1). 


© 2006 by Taylor & Francis Group, LLC 


Terminology of Insect Repellents 


41 


Rutgers 6-2-2 A repellent mixture consisting of 6 parts dimethyl phthalate, 2 parts Rutger’s 612, 
and 2 parts indalone), optimized for military 4 ' 11 use during World War II as M-250. 
semiocheniicals Chemicals involved in communication among organisms. 66 
SI units International System of Units (http://physics.nist.gov/cuu/Units/index.html). 
soluble Ability to dissolve in a given solvent, such as acetone, alcohol, water, 
solute That which dissolves, 
solution Solvent plus solute. 

solvent Liquid in which solute dissolves to form solution. 

specifications Standard descriptions of products for quality control purposes. For repellents and 
other pesticides, international specifications are prepared by the FAO and/or WHO, then 
adopted by the FAO/WHO Joint Meeting on Pesticide Specifications (JMPS, http://www. 
who.int/whopes/quality/en/), in conjunction with CIPAC analytical methods. Joint FAO/ 
WHO specifications are issued by the World Health Organization Pesticides Evaluation 
Scheme (WHOPES), available only in electronic format from http://www.who.int/whopes/ 
quality/en, providing a qualitative basis for production and procurement, 
spreader A chemical that increases the area that a certain volume of liquid will cover, 
sticker Something increasing adherence; formulation ingredient to enhance adherence of the 
active ingredient. 

stimulants Substances that cause insects to begin moving, copulating, feeding or laying eggs, 6 
hence qualified terms such as locomotor stimulant, mating stimulant and oviposition 
stimulant. The term feeding stimulant is synonymous with phagostimulant. 67 
substrate For purposes of repellents and other pesticides, the substrate is a treated surface (c.f. 

biochemical substrate—molecule acted upon by an enzyme; bioecological substrate— 
environment in which an organism lives). 

surfactant Chemical agent that increases the emulsifying, dispersing, spreading and/or wetting 
properties of another chemical when contacting a surface, 
suspension Finely divided solid particles mixed in liquid, in which they not soluble, 
synergist A substance that, when combined with another substance, gives effect that is greater than 
the sum of their individual effects. 

synomone Mutually beneficial signal chemical, released by members of one species, that affects the 
behavior of another species and benefits individuals of both species, 
synthetic Chemical compounds made by human directed process, as opposed to those of natural 
origin; the same material may be produced naturally or synthetically (e.g. PMD, 
Chapter 20). Since the 1940s (Chapter 1), most commercial repellents are synthetic 
compounds. Synthetic pyrethroids (q.v.) are important insecticides and irritant repellents, 
usually including a chrysanthemic moiety homologous to natural pyrethrins (q.v.). 
taxis Directional response to stimulus: movement towards the source being positive taxis; move¬ 
ment away from the source being negative taxis; c.f. kinesis. Chemotaxis (n.), chemotactic 
(adj.): movement responding to chemical (attractant or repellent), 
tolerance Having low susceptibility, due to high fitness of the individual or population; usually 
attributable to presence of some robust or resistant individuals from which a more obviously 
resistant population can be selected (due to increased frequency of resistant genotypes when 
successive generations are subjected to Darwinian selection). In many countries, regulatory 
systems set pesticide tolerances as maximum permissible levels of residues in foodstuffs etc. 
(established by EPA in the USA and by the ECB in the EU). Tolerance has special meaning 
for quality control purposes, whereby the permissible range of variation is defined in product 
specifications with respect to the active ingredient, e.g., mean+ 10%, possibly expressed 
as variance. Using this mathematical concept, Rutledge 63 64 assessed repellent tolerances of 
mosquito populations in order to compare ranges of responses and resistance potential. For 
pesticides generally, tolerance is recognized when the LC 50 of a population rises upto 5 
times greater than normal for a standard susceptible strain of the same species; higher ratios 


© 2006 by Taylor & Francis Group, LLC 


42 


Insect Repellents: Principles, Methods, and Uses 


(dose-efficacy comparisons between populations of the same species) indicate resistance 
(q.v.). 

toxics Based on the Greek word toxikon for arrow-poison, toxicology is the study of poisons 
biologically harmful substances and their effects; dose-dependent criteria allow any material 
to be toxic, serving as a toxicant or toxin for sensitive tissues or organisms, although this 
term is normally applied to hazardous pathogens, pesticides 68 and other chemicals; toxicity 
of pesticides is commonly measured (for each species) in terms of lethal concentrations or 
dosages at the 50% level (LCgo or LD 50 ) and the 99% level (LC 99 or LD 99 ) for comparative 
purposes when dealing with target insects and non-target species. The Toxics Release 
Inventory (TRI) is a publicly available EPA database (http://www.epa.gov/tri/) that contains 
information on toxic chemical releases and other waste management activities reported 
annually by industry and U.S. federal facilities. For chemical safety purposes, in setting 
tolerances (as above), toxicologists determine the ‘no observed adverse effect level’ 
(NOAEL) for laboratory animals. Mammalian toxicity values, required by regulatory 
authorities (such as the USEPA, Chapter 26) for assessing pesticides for regulatory 
approval, are based on effects of short-term (acute), long-term (chronic) and intermediate 
(sub-chronic) periods of exposure, as well as effects on development and reproduction, 
including mutagenicity and carcinogenicity, to establish dose-response relationships. For 
example, acute tests (so-called 6-pack) comprise oral, dermal and inhalation LDs, 
neurotoxicity, eye irritation, dermal irritation and sensitization (www.epa.gov/oppfeadl/ 
trac/a-toxreq.htm). The human equivalency potency factor (Q) is usually based on the oral 
exposure route, designated Q* when considered carcinogenic (www.epa.gov/pesticides/ 
carlist). The so-called Reference Dose (RfD) is the average daily oral exposure that is 
estimated to be unlikely to cause harmful effects during a lifetime. RfDs are generally used 
by the EPA for health effects that are thought to have a low threshold (dose limit) for 
producing effects. The International Programme on Chemical Safety 71 (IPCS) emphasizes 
the Acceptable Daily Intake 71 (ADI) for each chemical, aggregated from all sources of 
exposure, whereas the USEPA increasingly considers cumulative risk (www.epa.gov/ 
oppsrrdl/cumulative/) from exposure to groups of pesticides with an equivalent mode of 
action (e.g. organophosphates). Whereas the mode of action of insect repellents is not well 
understood (Chapter 11), the toxicology of repellent compounds is not difficult to assess by 
standard methods. 

U.K. United Kingdon of Great Britain and Northern Ireland, one of 25 Member States of the 
European Union, therefore subject to the Biocidal Products Directive (q.v.) for regulation 
of pesticides. 

USDA United States Department of Agriculture has a variety of Agencies, Offices and Services, 
notably the Agricultural Research Service (ARS) with long-term research on insect 
attractants and repellents. 

U.S. EPA, OPP United States Environmental Protection Agency, Office of Pesticide Programs 
(http://www.epa.gov/pesticides/), comprises several operating divisions, currently named: 
Antimicrobials, Biological and Economic Analysis, Biopesticides and Pollution Prevention, 
Environmental Fate and Effects, Field and External Affairs, Health Effects Division, 
Information Technology and Resources Management, Registration, Special Review and 
Reregistration. Collectively they are responsible for pesticide regulatory management in 
the USA. 

vapor pressure The property causing a chemical to evaporate, defined as the pressure of the vapor 
in equilibrium with the liquid or solid state; measured in joules, SI units of energy 
(International System of Units, http://physics.nist.gov/cuu/Units/index.html). 

vector Carrier of infection. Vector-borne pathogens cause disease; e.g., Plasmodium causes 
malaria, transmitted by vector Anopheles mosquito. 


© 2006 by Taylor & Francis Group, LLC 


Terminology of Insect Repellents 


43 


viscosity The property of liquids to resist flow, due to forces acting between the molecules. The SI 
physical unit of dynamic viscosity (greek symbol: p) is the pascal-second (Pa s), identical 
to 1 Ns/m 2 or 1 kg/(ms). 

volatility Rate of evaporation of liquid or solid. 

wetting agent A chemical that increases the liquid contact of dry material. 

WHOPES World Health Organization Pesticides Evaluation Scheme, responsible for assessments, 
specifications and recommendations for pesticides (including repellents) used for public 
health pest and vector control, 69 on behalf of Member States of the United Nations (U.N.). 
(http ://w w w .who.int/whopes/en/). 

zoophagy; zoophily Tendency of hematophagous insects to bite or prefer hosts other than humans 
(c.f. anthropophagy, anthropophily). 


References 

1. S. R. Christophers, Mosquito repellents, being a report of the work of the mosquito repellent inquiry, 
Cambridge 1943-5, J. Hyg., 45, 176, 1947. 

2. Webster’s Encyclopaedic Unabridged Dictionary of the English Language, New York: RHR 
Publishing, 2001. 

3. Chambers Twentieth Century Dictionary, Edinburgh: W & R Chambers, 1972. 

4. G. Gordh and D. Headrick, A Dictionary of Entomology, Wallingford: CABI Publishing, 2001. 

5. V. G. Dethier, Chemical Insect Attractants and Repellents, Philadelphia: Blakiston, 1947. 

6. V. G. Dethier, Man’s plague? Insects and Agriculture, Princeton, NJ: Darwin Press, 1976. 

7. V. G. Dethier et al.. The designation of chemicals in terms of the responses they elicit from insects, 
J. Econ. Entomol., 53, 134, 1960. 

8. H. K. Gouck, T. P. McGovern, and M. Beroza, Chemicals tested as space repellents against yellow- 
fever mosquitoes. I. Esters, J. Econ. Entomol., 60, 1587, 1967. 

9. C. E. Schreck, Spatial action of mosquito repellents, J. Econ. Entomol., 63, 1576, 1970. 

10. P. Granett, Studies of mosquito repellents, I. Test procedure and methods of evaluating test data, 
J. Econ. Entomol., 33, 563, 1940. 

11. P. Granett, Studies of mosquito repellents, II. Relative performance of certain chemicals and 
commercially available mixtures as mosquito repellents, J. Econ. Entomol., 33, 566, 1940. 

12. S. R. Christophers, Insect repellents, Brit. Med. Bull, 3, 222, 1945. 

13. S. R. Christophers, Mosquito repellents, Rev. Med. Mex., 26, 213, 1946. 

14. W. V. King, Repellents and insecticides for use against insects of medical importance, J. Econ. 
Entomol., 44, 338, 1951. 

15. W. V. King, Chemicals Evaluated as Insecticides and Repellents at Orlando, Florida, Bureau of 
Entomology and Plant Quarantine, Agriculture Research Service, United States Department 
of Agriculture. Agriculture Handbook 69, Washington, DC: U.S. Government Printing Office, 1954. 

16. USDA, Results of Screening Tests with Materials Evaluated as Insecticides, Miticides, and Repellents 
at the Orlando, Florida, Laboratory, April, 1942-April, 1947, Bureau of Entomology and Plant 
Quarantine, Agriculture Research Service, United States Department of Agriculture. Publication 
E-733, Washington, DC: U.S. Government Printing Office, 1947. 

17. USDA, Materials Evaluated as Insecticides, Repellents, and Chemosterilants at Orlando and 
Gainesville, FL, 1952-1964, Entomology Research Division, Agricultural Research Service, United 
States Department of Agriculture. Agriculture Handbook No. 340, Washington, DC: U.S. Government 
Printing Office, 1967. 

18. C. E. Schreck, K. Posey, and D. Smith, Repellent Activity of Compounds Submitted by Walter Reed 
Army Institute of Research, Part I. Protection Time and Minimum Effective Dosage against Aedes 
aegypti Mosquitoes, Agricultural Research Service, Technical Bulletin No. 1549, Washington, DC: 
United States Department of Agriculture in cooperation with Walter Reed Army Institute of Research, 
1977. 


© 2006 by Taylor & Francis Group, LLC 


44 


Insect Repellents: Principles, Methods, and Uses 


19. WHO, Equipment for Vector Control, 3rd ed., Geneva: World Health Organization, 1990. 

20. R.C. Muirhead-Thomson, The significance of irritability, behaviouristic avoidance and allied 
phenomena in malaria eradication. Bull. Wld. Hlth. Org. 22, 721, 1960. 

21. J. S. Kennedy, Behaviorally discriminating assays of attractants and repellents, in Chemical Control of 
Insect Behavior: Theory and Application , H. H. Shorey and J. J. McKelvey (Eds.), Chichester: Wiley, 
1977. chap. 13. 

22. J. L. Robertson and H. K. Preisler, Pesticide Bioassays with Arthropods, Boca Raton, FL: CRC Press, 
1992. 

23. WHO, Report of the WHO Informal Consultation on the Evaluation and Testing of Insecticides, Control 
of Tropical Diseases, Pesticide Evaluation Scheme, Informal Consultation, unpublished document 96.1, 
Geneva: World Health Organization, 1996. 

24. C. E. Schreck et al., Evaluation of personal protection methods against phlebotomine sand flies 
including vectors of leishmaniasis in Panama, Am. J. Trop. Med. Hyg., 31, 1046, 1982. 

25. EPA, Product Performance Test Guidelines, OPPTS 810.3700, Insect Repellents For Human Skin 
and Outdoor Premises, document EPA 712-C-99-369, December 1999, public draft issued 8 th June 
2006. following Scientific Advisory Panel (SAP) Meeting, April 7, 2000: Evaluation of Product 
Performance Testing Guidelines for Insect Repellents, SAP unpublished Report No. 00-02B, pp. 44- 
56; Federal Insecticide, Fungicide, and Rodenticides Act, Washington, DC, U.S. Environmental 
Protection Agency, SAP Report No. 2000-02, August 2, 2000. (http://www.epa.gov/OSA/hsrb/ 
backgrounddocuments.htm) 

26. P. M. Arguin, P. E. Kozarsky, and A. W. Navin, Health Information for International Travel. 2005- 
2006. Travelers’ Health: Yellow Book, Atlanta, GA: U.S. Centers for Disease Control and 
Prevention, 2005, Chap. 2. (http://www.cdc.gov/travel/yb/about.htm) 

27. CDC, Updated Information Regarding Mosquito Repellents, Fort Collins, CO: U.S. Department of 
Health and Human Services, Centers for Disease Control and Prevention, National Center for 
Infectious Diseases, Division of Vector-Borne Infectious Diseases, 2005. (http://www.cdc.gov/ 
ncidod/dvbid/westnile/RepellentUpdates.htm) 

28. E. A. Ross et al.. Insect repellant [sic] interactions: Sunscreens enhance DEET (/V,/V-diethyl-;n- 
toluamide) absorption, Drug Metab. Dispos., 32, 783, 2004. 

29. CTFA, International Cosmetic Ingredient Dictionary and Handbook, 10th ed., Washington, DC: 
Cosmetic, Toiletry, and Fragrance Association, 2004. (http://www.ctfa.org/Content/Navigation- 
Menu/CTFA_Books_CDs/Dictionary.htm) 

30. D. N. Roy, S. H. Ghosh, and R. N. Chopra, Comparative efficacy of different culicifuges under 
laboratory conditions. Parasitology, 34, 152, 1942. 

31. D. N. Roy and S. H. Ghosh, Further work on the comparative efficacy of different culicifuges under 
laboratory conditions. Parasitology, 34, 291, 1942. 

32. M. R. Berenbaum, The chemistry of defense: Theory and practice, Proc. Natl. Acad. Sci. U.S.A., 92, 
2. 1995. 

33. C. F. Curtis, J. Myamba, and T. J. Wilkes. Comparison of different insecticides and fabrics for anti¬ 
mosquito bednets and curtains, Med. Vet. Entomol., 10, 1, 1996. 

34. World Health Organization. IR3535; KBR3023; (RS)-methoprene 20%; pyriproxyfen 0.5% GR; and 
lambda-cyhalothrin 2.5% CS, Report of the Fourth WHOPES Working Group Meeting IR3535; 
KBR3023; (RS)-methoprene 20%; pyriproxyfen 0.5% GR; and lambda-cyhalothrin 2.5% CS, 
Document WHO/CDS/WHOPES/2001.2. Geneva: WHO, 2001. 

35. Y. Trongtokit et al.. Comparative repellency of 38 essential oils against mosquito bites. 
Phytotherapy Res., 19, 303, 2005. 

36. D. Barnard, Repellency of essential oils to mosquitoes (Diptera: Culicidae), J. Med. Entomol., 36, 
625, 1999. 

37. D. L. Kline et al., Olfactometric evaluation of spatial repellents for Aedes aegypti, J. Med. Entomol., 
40, 463, 2003. 

38. J. S. Kennedy, The excitant and repellent effects on mosquitoes of sub-lethal contacts with DDT. 
Bull. Entomol. Res., 37, 593, 1947. 

39. M. Coluzzi, Sulla irritabilita al DDT in Anopheles, Riv. Malariol., 42, 208, 1963. 


© 2006 by Taylor & Francis Group, LLC 


Terminology of Insect Repellents 


45 


40. E. J. Pampana, A Textbook of Malaria Eradication , 2nd ed., London: Oxford University Press, 1969, 
pp. 183-192. 

41. D. R. Roberts et al., A probability model of vector behavior: Effects of DDT repellency, irritancy, 
and toxicity in malaria control, J. Vector Ecol., 25, 48, 2000. 

42. E. B. Dogan and P. A. Rossignol, An olfactometer for discriminating between attraction, inhibition, 
and repellency in mosquitoes (Diptera: Culicidae), J. Med. Entomol., 36, 788, 1999. 

43. G. S. Fraenkel and D. L. Gunn, The Orientation of Animals, Oxford: Clarendon Press, 1940. 

44. Committee to Advise on Tropical Medicine and Travel (CATMAT), Advisory Committee Statement 
13, Public Health Agency of Canada, Statement on Personal Protective Measures to Prevent 
Arthropod Bites, Can. Commun. Dis. Rep., 31, ACS-13, 1-20, 2005 (http://www.phac-aspc.gc.ca/ 
publicat/ccdr-rmtc/05vol31/asc-dc-13/index.html). 

45. J. P. Grieco, A novel high throughput screening system to evaluate the behavioral response of adult 
mosquitoes to chemicals, J. Am. Mosq. Control Assoc., 21, 404, 2005. 

46. S. E. Lee, Mosquito larvicidal activity of pipernonaline, a piperidine alkaloid derived from long 
pepper, Piper longum. J. Am. Mosq. Control Assoc., 16, 245, 2000. 

47. I. M. Scott et al.. Botanical insecticides for controlling agricultural pests: Piperamides and the 
Colorado Potato Beetle Leptinotarsa Decemlineata Say (Coleoptera: Chrysomelidae), Arch. Insect 
Biochem. Physiol., 54, 212, 2003. 

48. I. M. Scott et al., Efficacy of Piper (Piperaceae) extracts for control of common home and garden 
insect pests, J. Econ. Entomol., 97, 1390, 2004. 

49. J. T. Arnason, T. Durst, and B. J. R. Philogen, Prospection d’insecticides phytochimiques de plantes 
temperees et tropicales communes ou rares, in Biopesticides d’origine vegetale, C. Regnault-Roger, 
B. J. R. Philogene, and C. Vincent (Eds.), Paris: Editions TEC and DOC, 2002, pp. 37-51. 

50. M. Miyakado et al., The Piperaceae amides, I: Structure of pipercide, a new insecticidal amide from 
Piper nigrum L, Agric. Biol. Chem., 43, 1609, 1989. 

51. M. Miyakado, I. Nakayama, and H. Yoshioka, Insecticidal joint action of pipercide and co-occurring 
compounds isolated from Piper nigrum L, Agric. Biol. Chem., 44, 1701, 1980. 

52. S. Dev and O. Koul, Insecticides of natural origin, Amsterdam: Hardwood Academic, 1997. 

53. I. M. Scott et al.. Insecticidal activity of Piper tuberculatum Jacq. extracts: synergistic interaction of 
piperamides, Agric. Forest Entomol, 4, 137, 2002. 

54. S.P. Carroll and J. Loye, A registered botanical mosquito repellent with deet-like efficacy., J. Am. 
Mosq. Control Assoc., 21, in press. 

55. C. F. Curtis, Control of Disease Vectors in the Community, London: Wolfe, 1990, pp. 79-80. 

56. F. Coro and S. Suarez, Review and history of electronic mosquito repellers, Wing Beats, 11,6, 2000. 
http://www.floridamosquito.org/WING/WBindex.html 

57. WHO, 7th Report of WHO Expert Committee on Insecticides, Geneva: World Health Organization. 
Technical Report Series, No. 125, 1957. 

58. J. Hemingway and H. Ranson, Insecticide resistance in insect vectors of human disease, Annu. Rev. 
Entomol., 45, 371, 2000. 

59. WHO, Vector Resistance to Pesticides, 15th Report of the WHO Expert Committee on Vector 
Biology and Control. World Health Organization, Technical Report Series, No. 818, 1992. 

60. WHO, Test Procedures for Resistance Monitoring in Malaria Vectors, Bio-efficacy and Persistence 
of Insecticides on Treated Surfaces, Report of the WHO Informal Consultation, document WHO/ 
CDS/CPC/MAL/98.12, Geneva: World Health Organization, 1998. 

61. D. W. Kolpin, Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. 
streams, 1999-2000: a national reconnaissance, Environ. Sci. Technol., 36, 1202, 2002. 

62. M. W. Sandstrom et al.. Widespread detection of /V,/V-diethyl-m-toluamide in U.S. streams: 
Comparison with concentrations of pesticides, personal care products, and other organic wastewater 
compounds, Environ. Toxicol Chem., 24, 1029, 2005. 

63. L. C. Rutledge et al., Studies on the inheritance of repellent tolerances in Aedes aegypti, J. Am. 
Mosq. Control Assoc., 10, 93, 1994. 

64. L. C. Rutledge, R. K. Gupta, and Z. A. Meher, Evolution of repellent tolerances in representative 
arthropods, J. Am. Mosq. Control Assoc., 13, 329, 1997. 


© 2006 by Taylor & Francis Group, LLC 


46 


Insect Repellents: Principles, Methods, and Uses 


65. H. J. Becker, The genetics of chemotaxis in Drosophila melanogaster: selection for repellent 
insensitivity, Mol. Gen. Genet., 107, 194, 1970. 

66. D. A. Nordlund, R. L. Jones, and W. J. Lewis (Eds.), Semiochemicals: Their Role in Pest Control, 
New York: Wiley, 1981. 

67. A. J. Thorsteinson, The experimental study of the chemotactic basis of host specificity in 
phytophagous insects, Canadian Entomologist, 87, 49, 1955. 

68. The WHO Recommended Classification of Pesticides by Hazard and Guidelines to Classification, 
International Programme on Chemical Safety. Geneva: World Health Organization, Geneva, 2004. 
http://www.inchem.org/pages/pds.html 

69. WHO, Pesticides and their Application for the Control of Vectors and Pests of Public Health 
Importance, 6th ed., WHO Department of Control of Neglected Tropical Diseases, and WHO 
Pesticides Evaluation Scheme (WHOPES), document WHO/CDS/NTD/WHOPES/GCDPP/2006.1, 
Geneva: World Health Organization, 2006. 

70. A. N. Gilbert and S. Firestein, Dollars and scents: commercial opportunities in olfaction and taste, 
Nat. Neurosci., 5(11), Supplement. Beyond the Bench: The Practical Promise of Neuroscience, 1045, 
2002 . 

71. International Programme on Chemical Safety, Inventory of IPCS and other WHO pesticide 
evaluations and summary of toxicological evaluations performed by the Joint Meeting on Pesticide 
Residues (JMPR), Evaluations through 2005, document WHO/PCS/06.2, World Health Organiz¬ 
ation, Geneva, 2005. 


© 2006 by Taylor & Francis Group, LLC 


3 


Vertebrate Chemical Defense: Secreted and Topically 
Acquired Deterrents of Arthropods 


Paul J. Weldon and John F. Carroll 


CONTENTS 

Introduction.47 

Arthropod Deterrents from Tetrapods.48 

Amphibians.48 

Snakes.51 

Birds.51 

Mammals.54 

Ungulates.54 

Humans.59 

Anointing.61 

Fumigation.64 

Discussion.65 

Acknowledgments.67 

References.67 


... for we know that the distribution and existence of cattle and other animals in South America 
absolutely depends on their power of resisting the attacks of insects: so that individuals which 
could by any means defend themselves from these small enemies, would be able to range into new 
pastures and thus gain a great advantage. 

(Darwin, 1857)' 


Introduction 

Arthropods profoundly affect the fitness of terrestrial vertebrates. Arachnids, centipedes, and insects 
opportunistically prey on small tetrapods. Some social hymenopterans—ants, bees, and wasps—fiercely 
defend their colonies against intruders, including a host of foraging vertebrates, via multiple stinging or 
biting attacks. Pelage- or plumage-degrading arthropods, such as lice, imperil their hosts by 
compromising the insulative and other qualities of the integument. The feeding activities of 
hematophagous insects, mites, and ticks irritate, weaken, and exsanguinate their victims. These 


47 


2006 by Taylor & Francis Group, LLC 




















48 


Insect Repellents: Principles, Methods, and Uses 


ectoparasites also transmit debilitating or lethal pathogens ranging from viruses to parasitic arthropods 
and inflict wounds vulnerable to microbial infection. Predatory, aggressive, and ectoparasitic arthropods 
are pervasive and potent agencies of natural selection that have forged in vertebrates an array of 
defensive adaptations. 

This chapter reviews evidence and suggestions that amphibians, snakes, birds, and mammals use 
chemicals in defense against arthropods. We focus on semiochemical effects rather than on mechanical 
protection afforded by secreted cuticles or adhesives. We also examine self anointing, where scent-laden 
materials are rubbed against the integument, and fumigation, which involves exposure of the integument 
to volatile compounds, as mechanisms by which tetrapods combat arthropods. 2-7 A number of functions 
have been proposed for topically acquired chemicals; most authors suggest that these substances thwart 
predators, ectoparasites, and/or pathogenic microbes. 7 

A variety of terms have been used to denote chemicals that affect nuisance arthropods (Chapter 2). 
A repellent, as Dethier et al. 8 proposed, denotes a chemical that elicits orientation away from its source 
(cf. Barton Browne 9 ). Thus, we confine our use of this term to cases in which arthropods avoid chemicals. 
We use “deterrent” to refer to any defensive chemical, including biocides, that reduces the risk of bodily 
harm. 10 


Arthropod Deterrents from Tetrapods 
Amphibians 

Amphibians harbor an array of bioactive compounds in their skin, including alkaloids, bufadienolides, 
and peptides. 11 The skin chemicals of a number of amphibians are believed to deter predatory and/or 
ectoparasitic arthropods. 

Field experiments in Costa Rica examined the acceptability of the poison frog Dendrobates pumilio 
and frogs of the genus Eleiitherodactylus (presumed to lack skin toxins) as prey to the large ant 
Paraponera clavata 12 and the ctenid spider Cupiennius coccineusP Ants typically refused to eat 
Dendrobates pumilio, retreating after touching frogs with their antennae, or releasing them after biting. 12 
Ants that bit Dendrobates pumilio typically wiped their jaws on their forelegs or tree bark. In contrast, 
ants readily attacked Eleutherodactylus spp., usually fatally. Similarly, the spider Cupiennius coccineus 
grasped and bit both kinds of frogs, but released Dendrobates pumilio and ate the Eleutherodactylus 
spp. 13 The tendency of these arthropods to reject Dendrobates pumilio after biting or antennating it 
suggests that they rely upon contact chemoreception to recognize this frog. 

Aquatic insects are major predators of amphibian larvae. Brodie and colleagues 14 ' 15 tested predatory 
larvae of the diving beetle ( Dytiscus verticalis), nymphs of the giant water bug ( Lethocerus americanus), 
and crayfishes ( Cambarus diogenes and Orconectes propinquus ) for responses to larval and metamorphic 
anurans ( Bufo americanus, Hyla crucifer, Rana clamitans, Rana palustris, and Rana sylvatica ) and 
urodeles ( Ambystoma maculatum and Notophthalmus viridescens) from the eastern United States. All 
larvae were attacked and consumed. Metamorphic stages of Bufo americanus, Rana palustris, Rana 
sylvatica, Ambystoma maculatum, and Notophthalmus viridescens, on the other hand, generally were 
rejected by the predators and survived. 

Histological studies of the skin of Rana sylvatica revealed that the unpalatability of this frog’s 
metamorphic stage correlates with the development of epidermal granular glands. 15 The glands are 
sparsely distributed in the larval stages and become abundant, larger, and appear active in late 
metamorphic stages. The greater immunity to predation of adults of this frog and other amphibians 
was attributed to the elaboration of defensive skin chemicals during maturation. 

Peterson and Blaustein 16 tested diving beetles ( Dytiscus sp.) and giant water bugs (Lethoceros 
americanus) for feeding responses to various developmental stages of a toad, Bufo boreas, and two frogs, 
Hyla regilla and Rana cascadae. These investigators failed to find evidence of stage-specific or species 


© 2006 by Taylor & Francis Group, LLC 




Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


49 


differences in the palatabilities of these anurans. Peterson and Blaustein attribute the discrepancies of 
their results with those obtained in other studies to possible species differences in chemical defenses or 
predator tolerances, or to problems in the design of previous investigations. 

Predators experienced with noxious amphibians often develop an aversion to them. This effect was 
demonstrated in the diving beetle (Dytiscus verticalis ). 17 Wild-caught diving beetle larvae were confined 
with eastern red-spotted newts (Notophthalmus viridescens), which are unpalatable to vertebrate 
predators 18 and leeches. 19 When tested later, beetle larvae rejected newts’ tails or aqueous extracts of 
newts absorbed onto cotton swabs. Thus, newts' skin chemicals acted as an aposematic cue. Starved 
beetles, however, seized newts more often than did beetles that had been fed meat. 20 Thus, although 
beetles were averse to (water-borne) chemicals from newts, their tendency to attack newts increased 
with hunger. 

Some frogs reside in ant nests, spider burrows or scorpion retreats where they access humid estivation 
cavities, refuge from predators, and/or prey (see Rodel and Braun 21 ). Skin chemicals are believed to 
protect these anurans from the arthropods with which they associate. Rodel and Braun, 21 for example, 
studied the frogs Kassina fusca and Phrynomantis microps, which live in nests of the ponerine ants 
Megaponera foetens and Pachycondyla tarsatus in the savanna of West Africa; Pachycondyla tarsatus is 
the largest African ant, and a hunter and scavenger. Humans were attacked when ant nests were 
excavated, but the frogs, whose burrows occur deeply within the nests, were unmolested. 

Encounters were staged between Pachycondyla ants and five frog species: the two species mentioned 
above that reside in ant nests; Hemisus marmoratus, which estivates underground (and may encounter 
ants); and Phrynobatrachus latifrons and Ptychadena maccarthyensis , both of which estivate above 
ground. Phrynobatrachus and Ptychadena were stung by ants and killed immediately, but the other frogs 
were unharmed. Phrynomantis microps assumes a crouched posture in the presence of ants and allows 
them to antennate, lick, and climb on its body (Figure 3.1). Ants that bit this frog wiped their mouthparts 
and antennae and the results of the encounter were fatal for one ant. 

The possible ant-deterrent properties of frog skin chemicals were examined by rubbing the ant- 
susceptible Phrynobatrachus latifrons against a live ant inquiline, Phrynomantis microps, and comparing 
ants’ responses to skin-rubbed versus untreated frogs. Untreated frogs generally were stung and killed; 
whereas only one of four Phrynomantis-tieatsd frogs was stung. In another experiment, termites 
(Macrotermes bellicosus) were dipped into water that had contained Phrynomantis frogs, and were 
then confined with ants. Ants stung both Phrynomantis-treated and untreated termites, but they attacked 
the latter more readily. Rodel and Braun 21 postulated that frogs are tolerated by ants because their skin 
contains a “stinging inhibitor.” Similarly, in Cameroon, the frog Kassina senegalensis resides in nests of 
the ant Megaponera foetens , and may rely upon mollifying chemicals, possibly mimics of ant 
pheromones, to do so. 22 

The “ant frog” ( Lithodytes lineatus), which ranges from the Peruvian Amazon to Surinam, typically 
inhabits nests of the leaf-cutting ant Atta ceplmlotes , 23 Wild-caught frogs emit a scent similar to that of 
the “maggi plant” ( Levisticum officinale), but captive-reared individuals and those held captive for 
several months lacked this scent. When placed near ant colonies, unscented frogs were attacked and 
killed, even by the colonies from which they were obtained. These observations imply that Lithodytes 
lineatus must emit special chemicals to enter ant nests unharmed. 

Williams et al. 24 tested the Australian sheep blowfly ( Lucilia cuprina) and the eastern goldenhaired 
blowfly ( Calliphora stygia) with skin secretions of the brown tree frog {Litoria ewingi). These blowflies 
are carrion feeders and do not pose a threat to frogs, but related flies parasitize frogs. An aqueous solution 
of Litoria skin secretions topically applied to third-instar larvae of Lucilia cuprina increased their 
mortality, but did not affect their behavior. When the skin solution was poured into glass vials containing 
adult Calliphora stygia, immersing their tarsi, flies vigorously groomed, exhibited “frantic” uncoordi¬ 
nated behavior, and became inverted. All flies died within 15 minutes. Calliphora stygia did not refuse 
food treated with frog secretions, but died after eating it. Williams et al. suggested that both physical and 
chemical properties contribute to the toxicity of Litoria skin secretions. Contact toxicity may be due to 
the occlusion of flies’ spiracles and the obstruction of respiration. 


2006 by Taylor & Francis Group, LLC 


50 


Insect Repellents: Principles , Methods, and Uses 



FIGURE 3.1 A Phrynomantis microps from Africa adopts a crouched posture while being examined by ants 
(Pachycondyla tarsatus). (From M.-O. Rodel and U. Braun. Biotropica, 31, 178, 1999. With permission.) 


Many frogs and toads emit distinctive scents, especially when disturbed. Smith et al. 25 suggested that 
volatile compounds discharged by anurans deter predators and ectoparasites. Williams et al. 26 
investigated whether the skin secretions of five Australian frogs —Litoria caerulea, Litoria splendida, 
Litoria rubella, Litoria rothi, and Uperoleia mjobergi —repel the mosquito, Culex annulirostris. 
A previous analysis of the parotoid gland secretions of Litoria caerulea had revealed P-caryophyllene, 
a presumed mosquito repellent. 27 Mosquitoes were confined in a choice chamber into which airsteams 
drawn over frogs’ skin secretions or water (control) were introduced, and the distribution of mosquitoes 
on either side of the apparatus was monitored. Mosquitoes avoided the frog-scented side the chamber 
only in response to the scents of Litoria rubella and Uperoleia mjobergi. The scent of Litoria caerulea 
was marginally repellent. 

To examine whether the skin chemicals of Litoria caerulea deter biting by mosquitoes, Williams 
et al. 26 applied an aqueous rinse of this frog’s skin secretions to the tails of mice (Mus musculus) and 
allowed mosquitoes to bite them; plain water was applied to control mice. Mosquitoes exhibited a greater 
latency (up to 50 minutes) to bite the secretion-treated tails than the controls. Analogous results were 
obtained with Litoria ewingi, the skin of which emits eucalyptol, limonene, and a-pinene. 28 The 
secretions of this frog applied to a human forearm delayed landing by Culex annulirostris for up to 
approximately 10 minutes. 


© 2006 by Taylor & Francis Group, LLC 




Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


51 


Snakes 

A number of reptiles, including snakes, lizards, and amphisbaenians, take shelter, feed, or deposit their 
eggs within insect colonies, somehow avoiding sustained attack. Chemicals seem likely to mediate these 
interactions, but explicit suggestions for this are rare. 

The Texas blind snake ( Leptotyphlops dulcis ) is a burrower that travels in columns of foraging army 
ants ( Neivamyrmex nigrescens) and feeds within ant colonies. When Leptotyphlops dulcis is manually 
placed into raiding columns of ants, it is attacked. 29 In response, it smears itself with feces and a clear 
viscous fluid discharged from its cloaca. 30 This exudate includes secretions from the cloacal scent glands, 
which contain a glycoprotein and C 12 -C 20 free fatty acids. 31 

The deterrent properties of the cloacal exudate of Leptotyphlops dulcis against ants were examined by 
wiping fluids off of a snake’s body, dissolving them in ethanol, and presenting them on the floor of an 
arena to Neivamyrmex nigrescens and other ants. 30 Ants placed at the juncture of areas treated with either 
snake cloacal extract or ethanol spent more time in the ethanol-treated area. Ants’ aversion to snake 
secretions was thought to be due to free fatty acids from the scent glands. 31 Gehlbach et al. 30 also 
observed that ants failed to attack Leptotyphlops dulcis maintained in laboratory ant colonies, suggesting 
that ant-derived chemicals acquired by snakes curtail attack. 

The Mexican short-tailed snake ( Sympholis lippiens) is a fossorial, insectivorous species from western 
Mexico that resides in colonies of leaf-cutting ants, Atta mexicana, which it eats. Sympholis possesses a 
tough skin thought to protect it from biting ants and an oily epidermis that is believed to deter them. 32 

Birds 

During the 1940s, Cott 33 undertook a series of field experiments in Egypt and Lebanon on chemical 
defense in birds, examining their acceptability to mammalian and insect predators (see Dumbacher and 
Pruett-Jones 34 ). While collecting bird specimens and preparing their skins, Cott noticed that the 
discarded carcasses of the laughing dove ( Streptopelia senegalensis ) were vigorously consumed by 
Oriental hornets ( Vespa orientalis), whereas pied kingfisher ( Ceryle rudis ) carcasses were ignored 
(Figure 3.2). Further comparisons of hornets’ responses to bird carcasses culminated in more than 140 



FIGURE 3.2 Oriental hornets ( Vespa orientalis) attack a freshly skinned carcass of a laughing dove ( Streptopelia 
senegalensis) (left), while ignoring the carcass of a pied kingfisher ( Ceryle rudis). (From H. B. Cott. Proceedings of the 
Zoological Society of London. 116, 371. 1947. With permission.) 


2006 by Taylor & Francis Group, LLC 



52 


Insect Repellents: Principles, Methods, and Uses 


experiments involving 38 bird species. Cott deemed hornets suitable as subjects because they convey 
what they consume back to their colony, thus their hunger state was presumed not to have varied 
substantially between experiments. 

Cott typically placed skinned pieces of two bird species side-by-side and observed hornets feeding on 
them until one was consumed. Tissues from different parts of birds varied greatly in their acceptability to 
hornets. The breasts and wings of Ceryle rudis, for example, were more acceptable than were the legs and 
visceral tissues. The breasts of other species also generally were preferred, in contrast to their legs which 
were consumed last or remained uneaten. Adipose tissue was uniformly avoided. Because different 
tissues varied in their acceptability, Cott compared responses by hornets to homologous tissues from 
different bird species. 

In addition to Ceryle rudis, the following birds were deemed unpalatable to hornets: the kingfisher 
(Alcedo atthis ), the blackcap (Sylvia atricapilla), the golden oriole ( Oriolus oriolus), the hoopoe ( Upupa 
epops), chats ( Oenanthe spp.), shrikes ( Lanius spp.), and swallows ( Hirundo spp.). Interestingly, hornets 
and domesticated cats (Felis catus ) exhibited similar feeding preferences for bird flesh. The 
unpalatability of these birds in Cott’s study implies that distasteful chemicals occur systemically 
because the specimens he presented were skinned. The responses of consumers to intact birds need to 
be assessed to evaluate this as a mechanism of avian defense. 

Thiollay 35 reported that the red-throated caracara (Daptrius americanus ), an insectivorous, forest¬ 
dwelling raptor ranging from Mexico to Brazil, emits volatiles that deter hymenopterans. He provided 
the following account of this bird attacking wasp nests: “As soon as one bird reached a nest, all the 
insects abandoned it and never attacked the raider, nor followed it when it earned the nest away. 
The wasps flew at a distance around the bird, rarely coming nearer than 1 m as long as it was on the nest. 
They returned to the remains of the nest shortly after the caracara left, sometimes within a few seconds 
.... The fact that wasps never attacked, nor even closely approached the caracaras raiding their nests, 
suggests the involvement of some powerful chemical repellent.” Daptrius possesses a bare, thin-skinned 
face and throat, a possible source of the putative wasp deterrents. 36 

Daptrius americanus is not consumed by humans in Guiana due to its disagreeable odor and taste. 35 In 
Mexico, this bird is avoided because a bluish dust on its feathers is believed to be poisonous. 37 Whether 
the chemicals from Daptrius thought by humans to be noxious are related to those that putatively deter 
wasps is unknown. Nonetheless, many birds are regarded by humans as characteristically malodorous or 
unpalatable, which may reflect their possession of deterrents against natural enemies. 38 

Tribespeople and field biologists in New Guinea consider the hooded pitohui (Pitohui dichrous ) to be 
unpalatable and noxious because individuals who handle it typically sneeze and experience numbness 
and a burning sensation. Dumbacher et al. 39 analyzed the feathers and other tissues of Pitohui dichrous, 
the variable pitohui (Pitohui kirkocephalus), and the rusty pitohui ( Pitohui ferrugineus). They discovered 
that the feathers and, to a lesser extent, the muscle tissue of these birds contain homobatrachotoxin 
(HBTX), a steroidal alkaloid first characterized from the skin of neotropical dendrobatid frogs (genus 
Phyllobates). HBTX is a potent neurotoxin that binds sodium channels and depolarizes electrogenic 
membranes. Further analyses demonstrated HBTX, batrachotoxinin-A, and other batrachotoxins in 
additional pitohui species and in an unrelated New Guinean bird, the blue-capped ifrita (Ifrita kowaldi). 40 
HBTX and other batrachotoxins have since been discovered in New Guinean melyrid beetles (Choresine 
spp.) that are consumed by pitohuis. 41 The Melyridae is a cosmopolitan family thought to be the dietary 
source of alkaloid toxins in both neotropical frogs and New Guinean birds. Alkaloids also occur in the 
feathers of the red warbler (Ergaticus ruber ) from Mexico, an insectivorous bird reputed to be inedible to 
humans. 42 

Dumbacher et al. 39 suggested that HBTX protects birds against predatory vertebrates, such as snakes. 
Other authors have postulated that this toxin combats ectoparasites. 43 ' 44 Poulsen 43 estimated the amount 
of HBTX in the skin of Pitohui dichrous, the most toxic pitohui, to be several orders of magnitude lower 
than that of the most toxic dendrobatid frog, Phyllobates terribilis. He surmised that higher toxin levels 
are necessary to deter vertebrates as formidable as those that threaten pitohuis, leaving ectoparasites as 
more likely targets. Mouritsen and Madsen 44 noted that batrachotoxins are toxic to a wide variety of 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


53 


insects, and they inferred that these compounds affect a spectrum of ectoparasitic arthropods. These 
authors also cited a survey of New Guinean birds in which pitohuis were found to harbor among the 
lowest tick loads of 30 passerine genera examined. 45 

Dumbacher 46 tested chewing lice for responses to the feathers of pitohuis and other birds. Lice were 
collected from 17 species of free-ranging birds in New Guinea, including pitohuis, and then confined in 
Petri dishes where they were given a choice between two contour feathers from different species. Lice 
presented with the feathers of pitohuis versus nonpitohuis, or the feathers of Pitohui dichrous versus 
those of the less toxic Pitohui cristatus, avoided the feathers of Pitohui dichrous in all cases. When 
exposed to the contour feathers from a pitohui and a nontoxic bird, louse mortality was higher with the 
pitohui feathers. For example, lice confined with feathers from Colluricincla megarhyncha , which was 
presumed to be nontoxic (cf. Dumbacher et al. 40 ), survived an average of 417 hours, but they did so for 
only 37 hours with the feathers of Pitohui dichrous. Louse species differed greatly in their tolerance of 
pitohui feathers. An undetermined species of Brueelia was the most sensitive; Neopsittaconirmus 
circumfasciatus was the most resistant. Dumbacher 46 postulated that lice that have coevolved with 
pitohuis are more tolerant of HBTX. 

Burtt 47 pointed out that the choice of feathers by lice in Dumbacher’s study, and the consequences of 
that choice on their survivorship, may have been influenced by feather microstructure peculiar to each 
bird species. He suggested that tests office from a single bird species would have improved Dumbacher’s 
experimental design. In addition to this concern, subsequent studies have revealed highly variable or 
undetectable levels of HBTX and the presence of other toxic compounds in pitohui feathers. 40 Thus it is 
unclear whether Dumbacher’s 46 results can be attributed solely to HBTX. Nonetheless, his study 
indicates that natural concentrations of chemicals from pitohui feathers adversely affect some lice. 

The effects of pitohui toxins on other arthropods need to be examined. Hippoboscid flies, which are 
hematophagous, often occur on pitohuis in nature. 46 Hippoboscids, along with mosquitoes and biting 
midges, transmit the pathogens that cause avian malaria, Haemoproteus, Leucocytozoon, and Plasmo¬ 
dium. One survey of these malarial parasites among birds of Australia and New Guinea revealed an 
infection rate for pitohuis that was similar to both the overall infection rate for their family, the 
Pachycephalidae, and the average prevalence of these parasites in New Guinea. 48 These results do not 
support the notion that toxins protect pitohuis against vectors of malarial parasites. Another survey of 45 
bird species from southeastern New Guinea, on the other hand, revealed that the Pachycephalidae, 
including pitohuis, had the lowest hematozoan loads, with no mature gametocytes in 21 individuals 
examined. 49 

The crested auklet ( Aethia cristatella) and the whiskered auklet ( Aethia pygmaea) are planktivorous, 
colonial seabirds from the Bering Sea and North Pacific Ocean. These birds emit a citrus-like aroma from 
their plumage that humans can detect emanating from their colonies. 50 Douglas et al. 50 ' 51 identified the 
following volatiles associated with this scent from the feathers of Aethia cristatella: n-octanal and, in 
lesser amounts, n-hexanal, n-decanal, (Z)-4-decenal, (Z)-4-dodecenal, (Z)-6-dodecenal, and hexanoic 
and octanoic acids. Hexadecanol, heptanal, nonanal, and decanal were identified from Aethia pygmaea. 
Douglas et al. 50 noted that similar aldehydes are used by heteropteran insects to repel predators, and they 
suggested that these compounds serve auklets as repellents of ectoparasitic arthropods. In addition, 
auklet volatiles were hypothesized to act as signals of mate quality related to the enhanced fitness 
associated with ectoparasite deterrence. 

Douglas et al. 51 tested the ticks Ixodes uriae (which parasitizes auklets) and Amblyomma americanum 
for responses to volatiles from the crested auklet. Laboratory tests of Amblyomma americanum nymphs 
were conducted by applying compounds to filter papers and attaching them to a heated rotating drum that 
served as an artificial host. Fewer nymphs transferred to papers treated with octanal or the mixture of 
auklet-derived volatiles than to control papers, and they spent less time on them. A field test indicated 
that Ixodes uriae nymphs were deterred by octanal and, to lesser extent, by decanal and a mixture of 
aldehydes. Ixodes uriae nymphs became moribund within 15 minutes after confinement with octanal; 
adults did so within one hour. 


2006 by Taylor & Francis Group, LLC 



54 


Insect Repellents: Principles , Methods, and Uses 


The suspicion that crested auklets in colonies near tundra habitats are vulnerable to mosquitoes 
prompted Douglas et al. 52 to test the effects of auklet volatiles on the yellow fever mosquito ( Aedes 
aegypti). Hexanal, octanal, (Z)-4-decenal. and hexanoic acid were applied separately or blended to filter 
papers laid on human hands that were then inserted into a chamber with mosquitoes; ethanol served as a 
control. All auklet-derived compounds deterred landing by mosquitoes. Octanal alone was as effective as 
the blend of volatiles and hexanal was as effective as hexenoic acid. Douglas et al. 52 postulated that 
volatiles emitted by auklets act against a broad spectrum of ectoparasitic arthropods. 

Investigations of the effects of auklet volatiles on lice, however, have yielded ambiguous or negative 
results. Lice of the genera Austromenopon and Quadraceps from wild-caught birds in Alaska became 
moribund within seconds after being confined with feathers treated with octanal or (Z)-4-decenal. 51 The 
louse loads of free-ranging crested auklets, however, were higher than those of the least auklet ( Aethia 
pusilla), which lacks a noticeable odor. 5 Moreover, when lice from the rock dove ( Columba livia), 
Campanulotes compar and Columbicola columbae, were confined with feathers from the crested auklet, 
least auklet, or rock dove, or were placed into beakers containing the carcasses of these birds, their 
survivorship did not differ between treatments. Douglas et al. 53 concluded that natural concentrations of 
crested auklet volatiles are not lethal to lice, but may repel or otherwise deter them. 

Most extant birds possess on their rump a uropygial gland which secretes an oil that is spread through 
the plumage during preening. Uropygial gland lipids inhibit microbial growth and thus retard feather 
degradation (see Moyer et al. 54 ). Poulsen 43 suggested that uropygial gland secretions also repel 
ectoparasitic arthropods. 

Moyer et al. 54 examined the effects of uropygial gland secretions from the rock dove on its host- 
specific feather lice, Campanulotes compar and Columbicola columbae. Uropygial gland oils applied to 
feathers obtained from glandectomized birds doubled the mortality of lice confined in jars, but louse 
loads on glandectomized versus intact birds did not differ. In fact, one dove that lacked a uropygial gland 
had among the lowest observed louse loads. Moyer et al. 54 discussed reasons for the disparate results of 
their in vitro and in vivo tests, including the possibility that excessive amounts of oil in their in vitro 
experiment killed lice by clogging their spiracles. These investigators entertained the prospect that 
uropygial gland secretions normally do not affect feather lice. 


Mammals 

Ungulates 

Investigations of the chemosensory responses by arthropods to mammals focus on attraction by 
ectoparasites to hosts. However, some studies reveal that chemicals from mammals deter these pests. 

Tsetse flies ( Glossina spp.) are hematophagous vectors of African trypanosomiasis. Blood-meal 
analyses of Glossina spp. in various parts of Africa reveal that they prefer particular mammalian hosts 
(see Galun 55 and Gikonyo et al. 56 ). Tsetse flies obtain a preponderance of blood-meals from a few 
ungulates, such as the warthog ( Phacochoerus aethiopicus), the bushpig ( Potamochoerus porcus ), the ox 
(Bos taurus), and the bushbuck ( Tragelaphus scriptus), but rarely attack other species, including the 
waterbuck ( Kobus defassa), the hartebeest ( Alcelaphus buselaphus), and the impala ( Aepyceros 
melampus ), even when these mammals are abundant. Nash 57 discussed why some mammals may be 
rejected as hosts by tsetse flies. He suggested that Glossina morsitans is averse to the scent of 
the hartebeest. 

Vale et al. 58 studied the responses of Glossina morsitans and Glossina pallidipes in Zimbabwe to ox 
urine and urinary phenols. Whole urine and some phenols attracted flies to traps or visual targets. 
2-Substituted phenols, on the other hand, suppressed attraction, an observation confirmed with Glossina 
pallidipes by Torr et al. 59 Species differences were observed in the synergistic effects of the phenols as 
attractants or deterrents. 58 4-Methylphenol (p-cresol) added to 3-«-propylphenol, for example, increased 
the trap captures of Glossina pallidipes, but reduced those of Glossina morsitans. 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


55 


Madubunyi et al. 60 investigated the responses of Glossina longipennis and Glossina pallidipes in 
Kenya to the urine of two preferred hosts, the African buffalo ( Syncerus coffer ) and domesticated 
cattle ( Bos taurus), and a nonpreferred host, the waterbuck. These investigators observed no differences 
in the trap catches with the urine of these animals, and they deemed it unlikely that tsetse flies locate 
hosts, or distinguish between hosts and nonhosts, by the scent of urine. Madubunyi et al. postulated 
that the chemicals used by tsetse flies to locate and distinguish among mammalian hosts arise from 
skin glands. 

Vale 61 conducted field studies in Zimbabwe to assess the responses of Glossina morsitans and 
Glossina pallidipes to various ketones, aldehydes, alcohols, carboxylic acids, and other compounds, 
some of which occur on the integument of mammals. A compound was characterized as “repellent” if it 
suppressed the number of flies attracted to a black cylinder model, which served as a visual target, placed 
near the release site of attractants, such as carbon dioxide or the scent of an ox. Acetophenone, which is 
emitted by bats, elephants, and other mammals, 62 and several related phenols; methyl ketones; 
aldehydes; and carboxylic acids, especially caproic acid, suppressed attraction. Glossina 
pallidipes was more effectively deterred by acetophenone and caproic acid than was Glossina morsitans 
(see also Torr et al. 59 on Glossina pallidipes ). A greater proportion of females than males of both species 
was deterred by acetophenone. 61 Acetic acid deterred both (biting) stomoxyine and nonbiting 
rnuscid flies. 

Gikonyo et al. 56 tested Glossina morsitans for responses to a preferred host, the ox, and a nonpreferred 
host, the waterbuck, in encounters and in feeding experiments using silicone membranes treated with the 
pelage extracts of these ungulates. No differences were observed in the tendencies of tsetse flies to land 
on live subjects or on extract-treated versus control membranes. However, flies that landed on a live 
waterbuck or membranes treated with its sebum changed probing sites more often, probed for longer 
periods, fed less frequently, and flew off sooner than did those landing on a live ox or membranes treated 
with ox sebum. The compounds from waterbuck to which tsetse flies attend were postulated to exhibit 
low volatility because the flies avoided waterbuck sebum after landing on treated membranes. However, 
tsetse flies that landed near but not on membrane zones treated with high doses of waterbuck sebum 
exhibited decreased feeding, thus indicating that they are capable of detecting volatile cues. 

Gikonyo et al. 63 examined the electroantennogram (EAG) responses of Glossina morsitans and 
Glossina pallidipes to volatiles extracted from absorbent pads impregnated with pelage chemicals from 
two preferred hosts, an African buffalo and an ox, and from the nonpreferred waterbuck. Glossina 
pallidipes reared in an insectary failed to respond to waterbuck volatiles, but field-trapped flies responded 
with EAG activity to 13 gas chromatographic peaks. Insectary-reared Glossina morsitans exhibited EAG 
responses to 14 components from waterbuck, and to 10 and 11 components from the ox and the buffalo, 
respectively. The following EAG-active compounds were unique to waterbuck (if also present in buffalo 
and ox, they occurred in trace amounts): S-octalactone, 2-methoxyphenol (guaiacol); 3-isopropyl-6- 
methylphenol (carvacrol); 2-octanone; 2-nonanone; 2-decanone; 2-undecanone; 2-dodecanone; and 
(£j-6,10-dimethyl-5,9-undecadien-2-one. 

Field studies by Vale and colleagues 58,61 in Zimbabwe demonstrated that guaiacol and C 4 -C 6 
methylketones reduced the trap captures of Glossina morsitans and Glossina pallidipes. Paradoxically, 
although a series of C 5 -C 9 straight-chain carboxylic acids unique to the waterbuck failed to elicit 
discernible EAG responses by tsetse flies in the study of Gikonyo et al., 63 two such compounds— 
pentanoic and hexanoic acids—suppressed attraction of flies to hosts in the field. 59 The various 
compounds present in the waterbuck blend, including the carboxylic acids, were postulated to 
differentially influence the distant- and close-range responses of tsetse flies to nonpreferred hosts. 63 
A notable result emerged from an interspecific comparison of flies’ EAG responses to waterbuck 
volatiles: Glossina pallidipes, which attacks waterbuck more readily than does Glossina morsitans, 
detected fewer methyl ketones of the repellent blend. 

Gikonyo et al. 64 tested Glossina morsitans to blends of EAG-active compounds from the ox, buffalo, 
and waterbuck in a choice wind tunnel. Compounds from the waterbuck included S-octalactone, 
carvacrol, m-cresol, C7-C10 aldehydes, and C8-C13 methlyketones. Tsetse flies exposed to waterbuck 


2006 by Taylor & Francis Group, LLC 



56 


Insect Repellents: Principles, Methods, and Uses 


volatiles failed to choose a consistent flight direction in the apparatus, as opposed to flies that embarked 
directly toward the scent of preferred hosts, and they came to rest in the odorless (control) arm of the 
tunnel. Flies exposed to waterbuck volatiles also flew shorter distances, made frequent and sharp in-flight 
turns, and fanned their wings against the tunnel walls—ostensibly attempting to fly out of the apparatus. 

Bett et al. 65 tested the efficacy of a synthetic blend of waterbuck volatiles in protecting oxen from 
Glossina pallidipes in Kenya. The artificial blend was applied to sachets and attached to oxen tethering 
posts. Flies were captured using electrified screens surrounding each bait animal. Feeding by Glossina 
pallidipes was reduced nearly 95% by the waterbuck blend. This blend also deterred Glossina 
swynnertoni, a species for which chemical control agents have been sought. 65 

Some mosquitoes also are deterred by the skin chemicals of bovids. Weldon 66 examined the responses 
of Aedes aegypti to pelage extracts of more than 15 ungulates, primarily artiodactyls. Mosquitoes were 
tested in a Plexiglas module using silicone feeding membranes, as described by Weldon et al. 67 Wells in 
the base of the module were filled with a 10% sucrose solution with added adenosine triphosphate and 
green food coloring. Mosquitoes were confined in chambers, the floors of which opened to allow them 
access to membranes placed over the wells. The number of mosquitoes landing on membranes treated 
with acetone or hair extracts was monitored for five-minute trials. The number of mosquitoes feeding was 
determined after each trial by crushing them on white paper towels and examining their remains for 
green fluid. 

Extracts of hair and sebum from the gaur (Bos frontalis), a large (650-1,000 kg) bovid with a greasy 
pelage that ranges from India to Indochina and the Malay Peninsula, significantly reduced landing and 
feeding by mosquitoes. Fraction-directed bioassays and analyses of gaur pelage extracts suggested that 
(6 R, 95, 105)-10-hydroxy-6,9-oxidooctadecanoic acid deterred landing and feeding by these mosqui¬ 
toes. 68 Ishii et al. 69 however, found that this compound, named 18-bovidic acid, exhibits the opposite 
stereochemical configuration, namely 65, 9 R, I UR. 

A three-dimensional structure-activity model designed to identify potential insect repellents indicated 
18-bovidic acid as a candidate compound. 70 Feeding studies using the silicone membrane feeding system 
described above confirmed that 18-(65, 9 R, 10f?)-bovidic acid, purified from the sebum of a gaur, deters 
landing and feeding by Aedes aegypti (Figure 3.3). 71 This compound contains a tetrahydrofuranoid ring 
that is adjacent to a hydroxyl group and is flanked by saturated hydrocarbon chains. This structure is 
reminiscent of that of acetogenins from custard apples (Annonaceae), which possess anti-feeding and 
biocidal properties against insects and other arthropods. 72 

Costantini et al. 73,74 studied mosquito attraction in Sudan using odor-baited entry traps, where air 
drawn over humans or cattle confined in tents was conveyed through tubing and released near a trap; 
carbon dioxide released in amounts comparable to those emitted by bait animals was released from 
control traps. Anopheles gambiae, an anthropophilic mosquito and a significant vector of malaria in 
Africa, exhibited an approximate 2:1 preference for human scent when compared to a human-equivalent 
of carbon dioxide, 73 but this species showed a nearly 20:1 preference for human scent when 
tested against the scent of cattle 74 (cf. Gillies 75 ). A similar effect was observed with Anopheles 
pharoensis. Costantini et al. 74 stated that these results may reflect mosquitoes’ avoidance of cattle as 
unsuitable hosts. 

Dekker and Takken 76 conducted a field study in South Africa on the attraction of mosquitoes to a 
human, a cattle calf, and carbon dioxide. Mosquitoes were captured in tents emitting the scents of these 
animals or carbon dioxide released in amounts comparable to those exhaled by them. The zoophilic 
mosquitoes Aedes mcintoshi. Anopheles coustani. Anopheles pretoriensis, and Anopheles rufipes were 
less attracted to humans than to a human-equivalent of carbon dioxide. Similarly, Culex quinquefas- 
ciatus, a species known to attack humans, preferred carbon dioxide over a live calf when this gas was 
released at a rate equivalent to that emitted by the calf. None of the Culex quinquefasciatus caught with 
the calf had fed, in contrast to those trapped in human-baited tents. These results imply that the 
chemosensory basis of host selection by mosquitoes involves not only attraction to preferred hosts, but 
avoidance of nonhosts. 


© 2006 by Taylor & Francis Group, LLC 


Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


57 


o 



o 

o 


1 mM 4mM 11 mM 33 mM acetone 

FIGURE 3.3 Mean percentages and 95% confidence intervals of Aedes aegypti landing (X) and feeding (•) on silicone 
membranes (9.6 cm 2 ) covering wells containing 10% sucrose solution and treated with 50 pL of acetone or 1, 4, 11, and 
33 mM solutions of 18-6S, 9R, 10^-bovidic acid (structure shown) in acetone. Fifty mosquitoes, five per trial, were observed 
for 5 min with each treatment. Values for landing were backtransformed, as described in Weldon et al. 7 An asterisk (*) 
indicates values significantly different from the control. (P. A. Evans, W. J. Andrews, and P. J. Weldon, Unpublished.) 


Dekker et al. 77 conducted wind-tunnel experiments on host-odor responses by Anopheles gambiae, 
which is anthropophilic, and Anopheles quadriannulatus, which feeds predominantly on bovids 
(cf. Pates et al. 78 ). Mosquitoes were given a choice of airstreams laden with either human or cattle 
scents emanating from nylon socks that had absorbed the skin secretions of these mammals, and they 
were captured in traps upwind. The resulting trap captures reflected mosquitoes’ typical host 
preferences: fewer Anopheles gambiae and Anopheles quadriannulatus entered airstreams containing 
cattle and human scents, respectively. Although cattle scent contains ammonia, which attracts 
Anopheles gambiae over a range of concentrations, more mosquitoes were caught in plain air traps 
than in those releasing cattle scent. Dekker et al. inferred that one or more compounds from cattle 
reduced the attractiveness of ammonia and possibly other cattle-derived volatiles. The deterrent effect 
of cattle scent on Anopheles gambiae also was observed in olfactometry tests by Pates et al. 79 

Observations of domestic cattle reveal variation within and between breeds in their attractiveness to 
ectoparasitic flies (see Jensen et al. 80 ). Jensen et al. 80 studied interactions between the horn fly 
(Haematobia irritans), an obligate, blood-feeding pest of pastured cattle in many parts of the world, 
and herds of Holstein-Friesian heifers in Denmark. Some individual heifers were highly attractive to 
flies, whereas others were fly-resistant. Heifers maintained their status with respect to fly-attractiveness 
over the two-year duration of the study. The exchange of three or four fly-resistant heifers for fly- 
susceptible individuals between herds of up to 17 cattle reduced overall fly loads for the herd, a pattern 
that was reversed when heifers were returned to their original herd. Thus, the degree to which flies 
menace a cattle herd depends upon the number of fly-susceptible and fly-resistant individuals it contains. 
The propensity of some individual cattle to draw fewer flies was attributed to their ability to emit 
chemicals that mask attractive cues. 81 

Birkett et al. 82 studied the role of volatiles from heifers in host selection by horn flies and face flies 
(Musca autumnalis) using gas chromatography-electroantennography (GC-EAG), GC-mass 


2006 by Taylor & Francis Group, LLC 



58 


Insect Repellents: Principles , Methods, and Uses 


spectrometry, EAG, and behavioral studies in the laboratory. Volatiles from heifers that exhibited high 
and low fly loads were analyzed and tested for EAG and behavioral activities with flies. Both fly species 
exhibited EAG responses to polar aromatic compounds, including phenol and o-cresol, m-cresol, and 
p-cresol, and to nonpolar aromatics, such as naphthalene and acenaphthene. In addition, horn flies 
exhibited EAG responses to the following volatiles from unattractive heifers: propylbenzene, styrene, 
camphene, 2-heptanone, and propyl butanoate. Phenol, m-cresol, and p-cresol, which were present in 
both attractive and unattractive samples, also were EAG-active. 

Compounds that exhibited EAG activity with an array of dipteran pests of cattle were presented to face 
flies in wind-tunnel experiments. 82 Propyl butanoate, a component of the least attractive heifer, 
suppressed attraction, as did naphthalene and, from urine, linalool. Interestingly, l-octen-3-ol and 
6-methyl-5-hepten-2-one, which were characterized in laboratory tests as attractants, tended to deter horn 
flies when artificially dispensed from free-ranging heifers. This result was attributed to the release of 
abnormally high levels of these compounds. The study of Birkett et al. demonstrates the importance of 
volatile cues in the differential attraction of hematophagous insects to individual bovids. Pickett and 
Woodcock 83 postulated that flies avoid individual cattle because they detect chemicals from them that 
reflect their immunological competence, thus rendering them less suitable hosts. 

Breeds of domestic cattle differ markedly in their vulnerability to ticks. Pan 84 suggested that reduced 
tick loads of Sahiwal versus lersey cattle are related to the greater production of sebum by the former 
breed. Other authors also have noted a possible correlation between sebum production and tick resistance 
among cattle breeds, e.g., Bonsma, 85 but it is unclear whether semiochemicals are involved. 

The brown ear tick ( Rhipicephalus appendiculatus ) and the red-legged tick ( Rhipicephalus evertsi) 
from Africa feed chiefly inside the ears and in the anal region of ungulates, respectively. Wanzala et al. 86 
observed that these ticks locate their characteristic feeding sites when placed on different parts of a host’s 
body. These investigators postulated that this ability involved concurrent responses to repulsive (from 
distant sites) and attractive (from feeding sites) cues. When tested in a climbing bioassay with extracts 
from different body regions of domestic cattle, brown ear ticks crawled toward ear volatiles and away 
from volatiles from the anal region. Conversely, red-legged ticks were attracted to anal volatiles and 
repelled by ear volatiles. The contrasting effects on ticks of chemicals from different body regions of their 
host represent a “push-pull” system of feeding site location, a phenomenon that may be widespread 
among organisms specializing on particular host microenvironments. 86 

The repulsive effects of host chemicals described by Wanzala et al. 86 facilitate the ticks’ location of 
their characteristic feeding sites, thus it is not clear if host chemical defenses are involved. Nonetheless, 
Sika 87 observed that when the anal scent of cattle was artificially applied to the area around their ears, 
brown ear ticks became disoriented, resulting in most subjects failing to locate their preferred 
feeding site. 

Many ticks exhibit an arrestant response (akinesis) to mammalian skin chemicals, a normal questing 
reaction in which ticks cease locomotion at ambush vantage points. Carroll and colleagues 88-90 observed 
that adult blacklegged ticks ( Ixodes scapularis) and American dog ticks ( Dermacentor variabilis) 
generally became akinetic in response to secretions from skin glands (tarsal and metatarsal) on the legs of 
the white-tailed deer ( Odocoileus viginianus ), but failed to do so in laboratory assays with some samples, 
avoiding them. These results were suspected to have been due to the contamination of some samples by 
urine, which mixes with glandular products during the deer’s rub-urination scent marking behavior. The 
tendency of blacklegged ticks to avoid interdigital gland secretions from the hindlegs of deer, but not 
those from the forelegs, is consistent with the hypothesized deterrent effect of urine. 90 

Host-seeking blacklegged ticks tested under conditions of high humidity (ca. 95% RH) avoided urine 
from bucks and nonestrous females of Odocoileus viginianus, but failed to do so at 50% RH. 91 Urine 
collected from the urinary bladder of a buck deterred ticks down to a 10,000-fold dilution. A subsequent 
study examining ticks’ responses to urine from immature and adult male and female deer revealed that 
only buck urine repelled them. 92 Carroll 91 suggested that fresh buck urine counteracts the arrestant 
properties of the glandular residues to which ticks ordinarily attend in identifying hosts. 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


59 


Humans 

Vale 93-96 and Hargrove 97 studied responses by tsetse flies (Glossina morsitans and Glossina pallidipes) 
in Rhodesia to humans and other mammals. These investigators observed that flies were averse to the 
scent of humans and less attracted to preferred hosts, such as the ox, when humans were nearby. This 
effect was especially pronounced in female flies, and thus could bias efforts to monitor tsetse fly 
populations when hand-netting was used to collect them. Vale 95 suggested that tsetse flies tend to avoid 
humans and other primates because they are vigilant and dexterous hosts that can capture and kill flies 
when they land. 

Vale 98 investigated the source and nature of human-derived deterrents of tsetse flies. Human body 
odor, but not human breath, significantly reduced fly catches in the field. When lactic acid, which recent 
studies show is distinctively abundant on human skin, 99 was applied to cloth, the catches of both Glossina 
morsitans and Glossina pallidipes were reduced. To further evaluate lactic acid as a deterrent, the 
number of flies attracted to an ox treated with 4.5 1 of a 1% solution of this compound were monitored; 
flies were examined to ascertain if they previously had fed. The catches of female Glossina morsitans and 
both sexes of Glossina pallidipes were reduced 50-66% for fed flies, but the catches of unfed flies were 
unaffected; too few male Glossina morsitans were caught to evaluate their response. With the exception 
of male Glossina morsitans, fewer flies engorged on lactic acid-treated versus untreated oxen. 

A number of investigators have reported that human skin secretions suppress attraction by mosquitoes 
or repel them. 100-104 Maibach et al. 103 observed that lipids from the elbow more effectively deterred 
Aedes aegypti than did lipids from the scalp. They inferred from this result that deterrent chemicals 
originate in the epidermis rather than in sebaceous glands, presumably because these glands are scarce or 
absent in the elbow region and abundant on the scalp. Thompson and Brown, 101 on the other hand, 
suggested that volatile acids released from the esterified components of sebum decrease the attractive¬ 
ness of human sweat to Aedes aegypti. Muller 104 observed that Aedes aegypti was attracted to sweat from 
the axilla, but avoided sweat from the trunk. 

Skinner et al. 105 used a dual-port olfactometer to compare the responses by Aedes aegypti to clean air 
or air laden with the extracts of human hands and elbows. Fewer mosquitoes landed near entry ports 
releasing volatiles from whole skin extracts and extract fractions obtained by thin-layer chromatography. 
Several nonpolar bands deterred mosquitoes, including one containing hydrocarbons; 106 the unsaturated 
components were deemed active. Squalene and a number of straight-chain alkanes and alkenes were 
presented to mosquitoes, but only 1-eicosene significantly deterred them. Neither this compound nor 
others tested singly, however, were as effective as the composite fraction. 

Another fraction from human skin extracts contained free fatty acids, 107 including saturated 
compounds that weakly deterred mosquitoes and at least two fractions containing unsaturated acids 
that were highly deterrent. Comparisons of mosquitoes’ responses to authentic saturated (C 5 -Ci 8 ) and 
unsaturated (C9-C20) compounds revealed a greater aversion to the latter. Further studies indicated three 
deterrent components in human skin extracts, the most abundant of which was lactic acid. 108 The other 
two components were tentatively characterized as hydroxy carbonyl compounds. 

Olfactometry studies by Bosch et al. 109 demonstrated that C] to C ]8 n-aliphatic carboxylic acids 
generally attract Aedes aegypti when combined with lactic acid, but this attractiveness was reduced with 
nonanoic and undecanoic acids; similarly, undecanoic and tetradecanoic acids reduced the attractiveness 
of lactic acid when combined with other carboxylic acids. Carboxylic acids generally attract Anopheles 
gambiae, as well, 110 but Smallengange et al. 111 observed that a mixture of twelve Ci-C ]6 carboxylic 
acids presented in an olfactometer repelled these mosquitoes. These results may have been due to 
impurities in the chemical samples. 

Lactic acid has been implicated in mosquitoes' responses to humans as both an attractant and repellent, 
depending upon its concentration and the responding species (see Steib et al." and Shirai et al. 112 ). 
Smith et al. 113 applied lactic acid to a cotton stocking at 3.56 mg/cm. 2 The stocking was placed on a 
human subject’s arm and inserted into a cage containing Aedes aegypti. Substantially fewer mosquitoes 
landed on the treated stocking than on an untreated one, and there were fewer bites. Subsequent 


2006 by Taylor & Francis Group, LLC 



60 


Insect Repellents: Principles, Methods, and Uses 


quantitative analyses of lactic acid on human skin revealed that the amount of this compound that had 
been applied to the stocking was 1,000-8,000 times greater than that normally present on human hands. 
Smith et al. concluded that the amount of lactic acid usually present on human hands attracts Aedes 
aegypti and probably never attains a concentration high enough to repel them. 

Shirai et al. 112 tested intact and proboscis-amputated Aedes albopictus with serial dilutions of lactic 
acid (1-10,000 ppm) applied to human forearms and hairless mice. Mosquitoes landed less frequently on 
humans and mice treated with both high and low concentrations of lactic acid than on water-treated 
controls. Intermediate concentrations (10-3,000 ppm) did not deter landing. Shirai et al. stated that the 
range of lactic acid concentrations they presented to mosquitoes was within that observed in human 
sweat, and that the minimum concentration of this compound found to deter mosquitoes was lower than 
that typically present on human skin. Shirai et al. suggested that the amount of lactic acid emitted by 
human sweat may reflect the circulating titers of this acid, and that mosquitoes shun hosts emitting large 
amounts of this compound to avoid imbibing it in blood-meals. 

A field study in Sudan by Costantini et al. 73 compared the attraction of mosquitoes to traps emitting 
human scent or carbon dioxide in amounts comparable to those emitted by a human subject. The number 
of Anopheles gambiae entering traps releasing human scent was twice that trapped with carbon dioxide 
alone, thus indicating that volatiles other than carbon dioxide are attractive to them. However, twice as 
many Anopheles pharoensis were captured in carbon dioxide-releasing traps as in the human-baited 
traps. This result might reflect the presence of human-derived inhibitors or repellents of this mosquito. 
Costantini et al., however, suggested that this may have been a spurious result, arising from variation in 
mosquitoes’ responses at the detection threshold of carbon dioxide used in their study. However, similar 
studies by Dekker and colleagues 76,77 on mosquitoes from South Africa have demonstrated that several 
zoophilic species avoid the scent of humans as nonhosts. 

A number of human-derived volatiles arise via microbial degradation of secretions from the apocrine 
glands of the axilla. Costantini et al. 114 investigated responses by strains of the mosquito Anopheles 
gambiae originating from East and West Africa to the axillary components, ( E )- and (Z)-3-methyl-2- 
hexenoic and 7-octenoic acids, using EAG and wind-tunnel and field-trapping bioassays in Burkina Faso. 
Both acids elicited EAG responses. In wind-tunnel experiments, fewer females entered chambers 
releasing carbon dioxide and a combination of a (£'/Z)-3-methyl-2-hexenoic acid isomeric mixture and 
7-octenoic acid when these compounds were presented in a range of doses. Field tests demonstrated that 
7-octenoic acid increased the catches of traps releasing carbon dioxide. However, when 7-octenoic acid 
was presented with (£'/Z)-3-methyl-2-hexenoic acid isomers, or when these isomers were presented 
singly or combined, fewer mosquitoes were captured in otherwise attractive traps. Costantini et al. 114 
stated that these acids may repel Anopheles gambiae or mask the attractiveness of carbon dioxide or other 
human scents. Alternatively, these investigators suggested that reduced captures with human-specific 
acids reflect the activity of these compounds in arresting the upwind flight of mosquitoes arriving at a 
scent source, a response that normally prevents them from overshooting their hosts. Thus, the inhibition 
of particular behaviors in the host-seeking repertoire of mosquitoes may, under some circumstances, 
facilitate host location. 

Humans exhibit individual variation in their attractiveness to mosquitoes (see Mukabana et al. 115 ). 
McKenzie 116 tested Aedes aegypti in an olfactometer to skin substances from individual subjects 
absorbed onto silicone membranes. The skin emanations from scent donors differed significantly in their 
attractiveness to mosquitoes. Some subjects were designated as “repellent.” Interestingly, the presence of 
highly attractive subjects appeared to render otherwise acceptable individuals less attractive. 

Bernier et al. 117 also tested Aedes aegypti in an olfactometer to chemicals from humans differing in 
mosquito attractiveness. An analysis of volatiles desorbed from glass beads handled by a less preferred 
subject revealed a greater abundance of the following compounds: 2-nonene; nonane; methyl undecane; 
pentacosane; decanoic acid; heptanal; 2,4-nonadienal; and nonanal. Olfactometry experiments demon¬ 
strated that these aldehydes inhibit mosquitoes’ normal attraction to lactic acid (Chapter 4) . 

Mukabana et al. 115 investigated the involvement of volatiles from human breath in the ability of 
Anopheles gambiae from Tanzania to distinguish between individual hosts. Mosquitoes were tested in 


© 2006 by Taylor & Francis Group, LLC 




Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


61 



FIGURE 3.4 A nest of wasps (Polybia diguetana) from Costa Rica is handled by a male farmer whose hand was covered 
with sweat from his underarm. (From A. M. Young, Biotropica, 10, 73, 1978. With permission.) 

an olfactometer with the breath, body odor, and total body emanations (both breath and body odor) of 
male subjects who differed in their attractiveness to mosquitoes. Breath was separated from body odors 
by requiring subjects confined in tents to mouth-breathe through a one-way valve, which diverted their 
exhalants away from other body effluents; air from an empty tent served as a control. More mosquitoes 
were captured in traps emitting body odors, total body emanations, and control odors than human 
breath, suggesting that breath contains deterrents. Moreover, when breath was removed from the scents 
of the subjects, they no longer differed in their attractiveness to mosquitoes. Mukabana et al. concluded 
that the body odors and total body emanations of humans have a kairomonal (attractive) effect on 
Anopheles gambiae and that human breath has an allomonal (repellent) effect on them. These authors 
suggested that the differential attractiveness of individual humans to mosquitoes is due to the effect of 
breath volatiles. 

The use of chemicals by mammals to deter hymenopterans has rarely been considered (but see 
Kingdon 118 on the African ratel, Mellivora capensis). Young 119 reported that the highly aggressive wasp 
(Polybia diguetana) from Costa Rica failed to sting and “seemed drugged” when its nest was handled by 
a male farmer who had coated his hand with his axillary secretions (Figure 3.4). The axillary odors of 
other farmers in Costa Rica and the United States also reportedly mollify wasps. Young 119 suggested that 
human sweat generally deters wasps. 


Anointing 

Many birds, primarily passerines, 120-133 and mammals, including insectivores, 134-135 primates, 4-5-136-143 
rodents, 144 carnivores, 3,4,145-150 and ungulates, 151 apply scent-laden materials to their integument. Birds 
hold objects in their beak and streak them through their plumage, primarily the wing feathers. Some 


2006 by Taylor & Francis Group, LLC 




62 


Insect Repellents: Principles , Methods, and Uses 


mammals, such as primates, rub materials into their fur with their forepaws. Others, like canids and 
felids, roll directly onto scent sources. Ungulates may acquire the scent of leaves by thrashing their horns 
in vegetation. 

Dolichoderine and formicine ants are used extensively in anointing by free-ranging birds. ’ In 
addition, birds rub themselves with other insects (see Whitaker ), millipedes (see Parkes et al. ), 
gastropods, 133 citrus 121 ' 130 ' 131 and other fruits, 124 ’ 133 onions, 122 resin, 120 and fresh 124 ' 129 and smoking 
vegetation. 123-125 Mammals also anoint with ants, 138 ' 143 citrus fruits, 4 ' 5 ’ 141 ’ 145 onions, 127142 milli¬ 
pedes, 139 ’ 140 ’ 142 and smoking vegetation. 136 in addition to noncitrus fruits, 137 leaves; 4 ’ 5 ’ 141 ’ 147 
resins; 149 toads; 135 carrion (see Reiger 146 ); and the feces, 134 ' 146 urine, 134 ’ 146 and skin gland secretions 
of heterospecifics. 144 ’ 146 ’ 148 

The presence of insecticides and other deterrents in the materials used in feather- and fur-rubbing 
points to these behaviors as mechanisms by which anti-arthropod compounds are acquired. 7 The 
increased incidence of anointing by animals confronted with heightened ectoparasite infestations also 
accords with this hypothesis. Free-ranging neotropical capuchin monkeys ( Cebus spp.), for example, rub 
their fur with leaves, fruits or millipedes most frequently during the wet season, when nuisance 
arthropods are abundant. 5 ’ 140 Similarly, many North American birds mb themselves with ants during 
molting periods that coincide with heavy rainfall; 128 the timing of this activity also is consistent with 
anti-microbial defense. 152 

Formic acid is believed to be the main compound appropriated by birds and mammals that anoint with 
formicine ants. In the 1940s, Dubinin observed that steppe pipits ( Anthus godlewskii) in Russia rubbed 
themselves with wood ants ( Formica rufa) (summarized in Kelso and Nice 126 ). Birds that rubbed 
themselves with ants, and those not observed to do so, were examined for feather mites ( Pterodectes 
bilobatus). The anointing birds harbored 87 dead mites and 612 mites that crawled over their feathers, 
171 of which died within 21 hours. Mites on nonanointing birds, on the other hand, remained attached to 
feathers. Of 758 mites collected from these birds, only seven died within 21 hours. Observations of 
hoopoes ( Upupa epops) in Tadzhikistan also suggested that mites ( Pterodectes cuculi ) are induced to 
crawl over feathers following bouts of anointing, but after 12 hours, only 1.7% of them died compared 
to up to 1.2% of the controls. Kelso and Nice 126 concluded that anointing with ants agitates ectoparasites, 
if it does not kill them. 

Eichler 153 placed lice that infested the feathers of domestic chickens (Callus domesticus), principally 
Eomenacanthus stramineus, into glass jars and sprayed them with a 50% solution of formic acid; controls 
were sprayed with water. Not surprisingly, all acid-sprayed lice died within a few minutes, whereas those 
treated with water survived. Wilson and Hillgarth 154 also observed that formic acid vapor kills lice and 
feather mites. Field observations of North American passerine birds 128 and studies under semi-natural 
conditions, 155 ’ 156 however, have failed to indicate that avian ectoparasites are affected by ant-derived 
fluids expressed during anointing. 

Millipedes, primarily those that secrete noxious benzoquinones, also are used for anointing by 
birds, 132 capuchin (Cebus spp.) 140 and owl monkeys (Aotus spp.), 142 and lemurs. 139 The benzoquinones, 
which millipedes release when disturbed, elicit fur-rubbing behaviors in capuchin 67 and owl monkeys. 142 
Valderrama et al. 140 observed that wedge-capped capuchin monkeys (Cebus olivaceous) in Venezuela 
rub themselves with the millipede, Orthoporus dorsovittatus, which secretes 2-methyl- 1,4-benzoquinone 
(toluquinone) and 2-methoxy-3-methyl-l,4-benzoquinone (MMB). These investigators hypothesized 
that benzoquinones acquired by monkeys repel mosquitoes. To test the plausibility of this hypothesis, 
Weldon et al. 67 presented toluquinone and MMB, individually and in a 1:1 mixture, to the mosquito 
Aedes aegypti on silicone feeding membranes placed over wells of human blood, a highly preferred 
food. Fewer mosquitoes landed on or fed through benzoquinone-treated membranes than did on solvent- 
treated membranes. Mosquitoes also exhibited higher flying scores when exposed to these compounds, 
a possible indication that they were repelled. 

Carroll et al. 158 tested lone star ticks (Amblyomma americanum) for responses to 1,4-benzoquinone, 
toluquinone, and MMB to examine the possible effects of these compounds in anointing by birds and 
mammals. Ticks typically spend hours or days wandering or feeding on their hosts, and thus are exposed 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


63 


over long periods to chemicals on their hosts’ integument. Carroll et al. observed that at low 
concentrations benzoquinones impaired ticks' ability to climb or right themselves, behavioral deficits 
that could affect their ability to access hosts. MMB also was mildly repellent. Ticks died when exposed to 
higher concentrations of these compounds. Thus, studies on mosquitoes and ticks show that the topical 
appropriation of benzoquinones may deter ectoparasites. However, key questions remain on the amounts 
of benzoquinones typically available to and appropriated by free-ranging animals. 

Citrus fruits are used in anointing by birds, 121 ' 130 ' 131 monkeys ( Cebus spp.), 4 ' 5 and canids. 145 After 
observing a grackle ( Quiscalus quiscula) rub its plumage with a slice of lime (Citrus aurantifolia ), Clayton 
and Vernon 130 tested the effects of volatiles from this fruit on feather lice ( Columbicola columbae ) from the 
rock dove. Lice confined with lime slices experienced higher mortality than did control lice with water. 
Tests of extracts from different parts of limes revealed that the biocidal compounds occur in the peel. 

The concentrated peel oils of citrus fruits are known to repel or kill a variety of insects, 159 including 
mosquitoes, such as Aedes aegypti, 160 Anopheles stephensi , 161 and Culex pipiens. 162 Weldon 163 
examined the effects of fresh unconcentrated lemon (Citrus limon ) peel extracts on Aedes aegypti 
using the membrane feeding system described above. Shallow incisions were made in the peel of 
organically grown lemons and the cut areas were lightly pressed against silicone membranes. 
The membranes were placed over wells containing a green-dyed sugar solution, and mosquitoes were 
allowed to land on and feed through them. Control membranes were treated with extracts of sliced kale 
(Brassica oleracea ) leaves or were left untreated (blank). The results of this experiment showed that 
mosquitoes landed and fed less on lemon-treated membranes than on either of the control membranes 
(Figure 3.5). Mosquitoes also flew more in chambers exposed to the lemon extract, suggesting that they 
were repelled by it. 

The peels of Citrus spp. are rich in volatiles, such as limonene, linalool, and citral, and nonvolatiles, 
such as coumarins and furanocoumarins. 159 ’ 164 These compounds are known to repel 165 or kill 166 



lemon kale blank 

FIGURE 3.5 Mean percentages and 95% confidence intervals of Aedes aegypti landing (X), flying (A), and feeding (•) 
on silicone membranes (9.6 cm 2 ) covering wells containing 10% sucrose solution and treated with a sliced lemon (Citrus 
limon ) peel or kale (Brassica oleracea ) leaves or left untreated (blank). Sixty mosquitoes, five per trial, were observed 
for 5 min with each treatment. Values for landing and flying were backtransformed, as described in Weldon et al. 67 
The values of all measures for lemon differ significantly from those for kale and blank conditions. (P. J. Weldon, 
Unpublished.) 


2006 by Taylor & Francis Group, LLC 



64 


Insect Repellents: Principles , Methods, and Uses 


a variety of arthropods, including vertebrate ectoparasites such as fleas, 167 ' 168 lice, 169 and mites. 170 
Studies are needed to assess whether birds and mammals are protected from nuisance arthropods by 
topically appropriating these compounds. 

The propensity of domestic cats to rub and roll on the leaves of catnip (Nepeta cataria ) is well known 
(see Tucker and Tucker 147 ). This response is elicited by nepetalactone, a cyclopentanoid monoterpene 
that protects Nepeta spp. from phytophagous insects. 171 Nepetalactone also repels other types of insects, 
including mosquitoes. Petersen 172 tested Aedes aegypti in a static-air choice test apparatus with catnip 
essential oil and Z,ii-nepetalactone, both of which were avoided. Similar results have been reported in 
other studies with catnip essential oil 172 ' 174 and nepetalactone. 174 E,Z- and Z,ii-nepetalactone, presented 
singly or in a 1:1 mixture, reduced feeding by Aedes aegypti in tests with both artificial feeding 
membrane and human subjects. 174 The deterrent properties of nepetalactone demonstrated in these 
studies support an anti-ectoparasite function for the catnip (anointing) response, thus providing a 
plausible alternative, if not more compelling, interpretation of this felid behavior than that of displaced 
socio-sexual or hallucinogenic responses (see Tucker and Tucker 147 ), or as a nonfunctional activity. 175 

Kodiak bears ( Ursus arctos ) chew “bear root” ( Ligusticum wallichii) and rub themselves with a 
mixture of the root juice and saliva. 150 Passreiter et al. 176 observed that the essential oil of Ligusticum 
mutellina from central and southern Europe kills third instar armyworms ( Pseudaletia unipunctata ) when 
topically applied to them, and that it contains phenylpropanoid insecticides, such as ligustilide. The 
effects of chemicals from Ligusticum spp. on the ectoparasites of ursids should be examined. 


Fumigation 

In fumigation, animals are exposed to volatile (usually plant-derived) chemicals that deter microbial 
pathogens and/or ectoparasites. Fumigation has been most extensively studied in birds that add fresh 
aromatic leaves to their nests, usually in amounts small relative to the structurally supportive nest matrix. 
Phytochemicals involved in nest fumigation are thought to repel ectoparasites or disrupt their 
reproduction or development (see Clark - ; cf. Fauth et al. and Gwinner et al. ). 

Surveys of falconiforms and passerines reveal that birds that re-use nests, ' or build nests in 
enclosed spaces 180 —situations where the risks of parasite or pathogen loads are high—are more likely to 
use fresh nest vegetation than are those that infrequently re-use nests or use open, cup-like nests. 
Comparisons of the plants locally available with those incorporated into nests reveal that birds select 
particular aromatic species for fumigation. House sparrows ( Passer domesticus) in India, for example, 
use the leaves of the margosa (neem) tree ( Azadirachta indica), ignoring many other available plants. 181 
The leaves of this tree contain numerous biocidal compounds, including (3-sitosterol, a repellent and 
oviposition inhibitor of mites. 

European starlings (Sturnus vulgaris ) in the eastern United States select the following plants for their 
nests: agrimony (Agrimoniaparx’iflora), wild carrot (Daucus carota), fleabane ( Erigeron philadelphicus), 
yarrow ( Achillea millefolium ), purple dead-nettle ( Lamium purpureum), and goldenrod ( Solidago 
rugosa). ’ “ Chemical analyses of these plants reveal mono- and sesquiterpenes, including carene, 
cymene, limonene, myrcene, ocimene, a- and [3-pinene, a-phellandrene, sabinene, a-terpineol, and 
a-terpinoline, many of which are known to repel or kill insects and other arthropods. 165 166 Goldenrod 
contains 2-bornyl acetate and farnesol, which are analogs of juvenile hormone. These compounds suppress 
molting in arthropods, and thus may interfere with the growth and reproduction of ectoparasites. 

Clark and Mason 180 tested the effects of nest fumigation by starlings on the hatching success of lice 
(Menacanthus sp.) and on the survival of adult northern fowl mites ( Ornithonyssus sylviarum). Volatiles 
from plants preferred by starlings more effectively retarded the hatching of lice than did volatiles from a 
random sample of nonpreferred plants, but mite survival was not affected. Nests devoid of wild carrots, 
however, possessed larger mite populations than those containing this plant. 180 In the laboratory, the 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


65 


emergence of feeding mite instars from nests containing wild carrot and fleabane, both preferred by 
starlings, was suppressed, whereas garlic mustard ( Alliaria officinalis), a commonly available 
nonpreferred plant, had no effect. 

Behavioral and electrophysiological studies demonstrate that birds use olfaction to identify fumigation 
plants. “ Clark and Smeraski showed that the olfactory sensitivity of European starlings varies 
seasonally, peaking during spring, when birds are breeding and building nests, and waning when birds 
enter a nonbreeding condition. Seasonal differences were demonstrated in birds’ responses to a variety of 
olfactory stimuli, including both preferred (dead-nettle and wild carrot) and nonpreferred (garlic 
mustard) nest plants. The seasonal expression of heightened olfactory sensitivity in starlings may 
allow them to identify fumigating plants. Clark and Mason 182 hypothesized that birds select plants based 
upon of their aromaticity and chemical complexity, rather than upon the presence of 
particular compounds. 

Cavity-nesting blue tits ( Parus caeruleus ) on the island of Corsica build nests that contain aromatic 
plants, including the leaves of Achillea ligustica and Lavandula stoechas. lM Birds regularly replace 
these leaves in their nests between the onset of egg laying and chick fledging. The postulated antiseptic 
and insect repellent function of leaf placement was corroborated by the presence of leaf volatiles such as 
camphor, eucalyptol, limonene, linalool, myrcene, piperotenone, and terpin-4-ol, which are bacterio¬ 
static and/or insect deterrents. 165 ' 166 

Nestlings lack protective feathers and are thus vulnerable to hematophagous insects. Lafuma et al. 186 
investigated whether aromatic plants in the nests of blue tits repel the mosquito, Culexpipiens, a vector of 
avian malaria. Mosquitoes were placed into boxes containing leaves and allowed to escape through tubes. 
The mixture of plants repelled them, but only the leaves of Achillea or Lavandula did so alone. 
Mosquitoes allowed to choose between domestic chicks confined in boxes with or without aromatic 
leaves fed less on chicks with the leaves of Achillea or Lavandula. Lafuma et al. suggested that leaf 
volatiles repel mosquitoes or mask the odors of their hosts. 

Some mammals construct nests that include aromatic leaves or bark. Hemmes et al. 187 observed that 
the dusky-footed wood rat ( Neotoma fuscipes) in the western United States deposits fresh foliage, 
including the leaves of California bay ( Umbellularia californica), oak ( Quercus spp.), and toyon 
(Heteromeles arbutifolia), in or near their stickhouses. Bay leaves, in contrast to other foliage, were 
deposited by wood rats more often near nests than away from them, and were nibbled, leaving shallow, 
sporadic lacerations along leaf margins. This unique nibbling pattern suggests that bay leaves are not 
used for food, but for the release of volatiles. When larvae of the cat flea (Ctenocephalides felis) were 
incubated with intact or tom leaves from bay, oak, or toyon, or without foliage, their survival was more 
severely reduced with torn bay leaves (27%) than with the other conditions (88-94%). This study 
suggests that volatiles from bay leaves reduce nest-borne ectoparasites of Neotoma fuscipes. Studies are 
needed of the effects of volatiles from mammalian nest materials on other ectoparasites. 

Fumigation may be the mechanism by which animals are tolerated entering or nesting close to social 
insect colonies. The Tui parakeet ( Brotogeris sanctithomae), the cobalt-winged parakeet ( Brotogeris 
cyanoptera), and the black-tailed trogon ( Trogon melanurus) in the Peruvian Amazon, for example, nest 
in arboreal termitaria inhabited by both termites and an aggressive biting ant, Dolichoderus sp. 188 
Brightsmith 188 suggested that birds acquire the strong smell of Dolichoderus. thereby chemically 
camouflaging their nest and avoiding attack. Similarly, Janzen 190 suggested that the nests of birds 
built in swollen-thorn acacias of Central America and northern Colombia acquire the scent of ants, 
probably trail substances, and thus permit birds to avoid being evicted by ants. 


Discussion 

A taxonomically and ecologically diverse array of tetrapods produces or appropriates arthropod 
deterrents, borne principally on the integument. The skin of vertebrates has long been known as 


2006 by Taylor & Francis Group, LLC 



66 


Insect Repellents: Principles , Methods, and Uses 


a source of novel natural products featuring diverse chemical classes and structures, e.g., unusual 
branching patterns and unsaturation sites, that contrast with those of compounds from internal tissues. 190 
The unique chemistry of the skin has variously been attributed to the production of antibiotics, 
pheromones, or water retardants. We propose that the chemical diversity of the vertebrate integument, 
and interspecific variation in skin chemical profiles (see Weldon 191 ), are related, in part, to the 
elaboration of arthropod deterrents. Indeed, some vertebrate integumentary glands may function to 
secrete compounds that combat arthropod pests. The deterrence of biting flies by sebum, 56 ’ 63 ’ 71 ’ 101 for 
example, suggests an important but underappreciated role for mammalian sebaceous glands, the function 
of which has long been unclear. 192 

Many compounds identified from the skin of tetrapods are known to adversely affect arthropods. The 
odoriferous pelage of the reticulated giraffe (Giraffa Camelopardalis reticulata), for example, emits 
indole and skatole, 193 which deter some mosquitoes ( Aedes spp.). 194 195 Another giraffe pelage 
compound, p-cresol. is a pheromone for many ticks, 196 but in high concentrations (comparable to 
those arising from giraffes) it repels them. 193 Tests of chemicals known to occur on the skin of tetrapods 
have revealed additional compounds that deter ectoparasites. 51 ' 52 ' 61 ' 109 ’ 114 Even some free fatty acids, 
which are ubiquitous tetrapod skin products, are insecticidal, 166 and thus may discourage 
some ectoparasites. 

Anointing and fumigation enable tetrapods to acquire chemicals from a variety of exogenous sources. 
Key questions, however, remain regarding the defensive significance of topically appropriated 
chemicals. Although laboratory studies have demonstrated that compounds present in anointing and 
fumigation materials deter some ectoparasites, studies are needed on the quantities of chemicals that are 
topically acquired and the extent to which they deter arthropods under natural conditions. Formic acid, 
for example, kills feather lice in laboratory tests, 153 ’ 154 but neither controlled aviary experiments 155 ' 156 
nor observations of free-ranging North American birds 128 support the hypothesis that ant secretions 
topically applied to the plumage reduce louse loads. These results have prompted some authors to dismiss 
anointing as a mechanism of ectoparasite control. 128 However, taxon-specific ectoparasites, such as 
feather lice, may be more tolerant of their host’s defensive chemicals if they and their hosts have 
coevolved. 34 ' 46 Hypotheses on the defensive function of chemicals should be tested with a spectrum of 
actual and potential ectoparasites to consider possible species differences in chemical tolerance, host 
dependence, and other variables. 

An arsenal of chemicals, including fumigant repellents, contact irritants, neurotoxins, masking agents, 
and hormonal antagonists/mimics, are thought to be deployed against arthropods by amphibians, reptiles, 
and mammals. The nature of the responses elicited by these compounds depends upon their 
concentrations, 56,58 ’ 64 ’ 112 ’ 113 the duration of exposure arthropods receive to them, 24 ’ 46 ’ 167 ’ 168 synergism 
of components of a blend. 58 ’ 64 and the sex, 58 ’ 93-96 ’ 98 rearing conditions, 63 ’ 64 and hunger state of the 
arthropods, 20 ’ 56 ’ 96 ’ 98 to name a few variables. Future studies examining species and populational 
differences in the ectoparasites’ responses to vertebrate-derived chemicals, including deterrents, may 
provide useful information for integrated pest management, in addition to elucidating microevolutionary 
variation in host-chemical preferences. 

Pickett and Woodcock 83 suggested that hematophagous insects avoid individual cattle because they 
emit volatile chemicals that denote their immunological competence. Biting flies also avoid the scents of 
nonhost species, ’ ’ ’ and it may be commonplace for foraging arthropods to do so. Aside 

from circumventing immunological defenses, ectoparasites should be selected to avoid chemicals 
denoting organisms that are dangerous, noxious or otherwise unprofitable as hosts. Conversely, would- 
be hosts should be selected to signal their status as an inappropriate resource, thus discouraging encounters 
with potentially injurious ectoparasites. Eisner and Grant 198 suggested that "olfactory aposematism,” 
where chemicals advertise an organism’s distasteful or other undesirable qualities to predators, is 
commonplace. Chemical aposematism vis-a-vis ectoparasites may also be widespread. 

Amphibians, reptiles, and mammals exhibit a variety of behavioral, physiological and anatomical 
defenses against nuisance arthropods. We have focused on their potential use of chemical defenses and 
described how these agents may reduce vulnerability to predators, ectoparasites, and disease vectors in an 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


67 


attempt to frame questions on this expanding area of interest. What roles do chemicals play in the 
symbiotic and other intimate associations between tetrapods and social insects? ’ ' ' Can the 

reduced intensity of ectoparasitic infestations 44,45 ' 55-57 ' 80,82 and the lowered incidence of vector-borne 
diseases among free-ranging animals 49 point to organisms profitably investigated for arthropod 
deterrents, or be used to assess the efficacy of these defenses in nature? 48 How have tetrapods converged 
with plants and invertebrates in the chemicals they use to counter arthropods? 50,51,165 Collaborations 
between chemists and biologists—a hallmark of the discipline of chemical ecology—are needed 
to address these questions, ideally, with the combined participation of arthropod and 
vertebrate specialists. 

Acknowledgments 

W. J. Andrews; U. R. Bernier; M. S. Blum; J. F. Butler; E. D. Brodie, Jr.; R. L. Chazdon; D. H. Clayton; 
P. Coon; M. Debboun; H. D. Douglas, III; T. Eisner; J. A. Endler; P. A. Evans; T. Falotico; R. C. 
Fleischer; H. W. Greene; A. Hassanali; R. Heyer; P. Holm; M. B. Isman; A. F. Jahn; S. Krane; M. 
Kramer, N. Kreiter; P. Manly; J. Millar; J. A. Pickett; M.-O. Rodel; B. P. C. Smith; T. F. Spande; D. 
Strickman; J.-M. Thiollay; R. K. Vander Meer; G. B. White; and A. M. Young provided materials, 
information, or comments on the manuscript. The Association for Tropical Biology and Conservation 
and the Zoological Society of London granted permission to reproduce figures. J. Bogard (American 
Council on Education, Washington, D. C.); A. Hutchinson; P. Lasker; M. Rosen; D. T. Steere, Jr. 
(Smithsonian Libraries, Washington, D. C.); M. Lothers; W. Olson; W. Thompson; C. Toefield Keen; F. 
Tyler (National Agricultural Library, Beltsville, Maryland); C. Twose (William H. Welch Medical 
Library, Johns Hopkins University, Baltimore, Maryalnd); and H. Brooks (Gorgas Memorial Library, 
Walter Reed Army Institute of Research, Silver Spring, Maryland) supplied references. L. Morel and M. 
Olshausen provided translations. Preliminary studies of mosquitoes by P. J. W. were supported by the 
Chemicals Affecting Insect Behavior Laboratory, USDA, Beltsville, Maryland. J. P. Benante; R. 
Coleman, W. Dheranetra, M. Dowler, S. Gordon, L. Jones, N. McLean-Cooper, E. Rowton, and J. 
Williams facilitated mosquito studies in the Department of Entomology, Walter Reed Army Institute of 
Research, Silver Spring, Maryland. D. L. Armstrong (Henry Doorly Zoo, Omaha, Nebraska), J. Chatfield 
(Gladys Porter Zoo, Brownsville, Texas), and J. Zeliff (Silver Springs Animal Park, Silver Springs, 
Florida) provided samples from animals in their care. This chapter was written while P. J. W. was 
supported by Bedoukian Research Inc., Danbury, Connecticut. 

References 

1. C. Darwin, The Origin of Species by Means of Natural Selection, or the Preservation of Favoured 
Races in the Struggle for Life, London: John Murray, 1859. 

2. L. Clark, The nest protection hypothesis: The adaptive use of plant secondary compounds by European 
starlings, in Bird-Parasite Interactions, J. E. Loye and M. Zuk (Eds.). Oxford: Oxford University 
Press, 1991, p. 205 

3. D. H. Clayton and N. D. Wolfe, The adaptive significance of self-medication, Trends in Ecology and 
Evolution, 8, 60, 1993. 

4. J. P. Berry, M. A. McFerren, and E. Rodriguez, Zoopharmacognosy: A “biorational” strategy for 
phytochemical prospecting, in Phytochemicals and Health, D. L. Gustine and H. E. Flores (Eds.), 
Rockville. MD: American Society of Plant Physiologists, 1995. p. 165. 

5. M. Baker. Fur rubbing: Use of medicinal plants by capuchin monkeys ( Cebus capucinus), American 
Journal of Primatology, 38, 263, 1996. 

6. B. L. Hart. Behavioural defence, in Host-Parasite Evolution: General Principles and Avian Models, 
D. H. Clayton and J. Moore (Eds.), Oxford: Oxford University Press, 1997. p. 59. 

7. P. J. Weldon. Defensive anointing: Extended chemical phenotype and unorthodox ecology, 
Chemoecology, 14, 1, 2004. 


2006 by Taylor & Francis Group, LLC 



68 


Insect Repellents: Principles , Methods, and Uses 


8. V. G. Dethier, L. Barton Browne, and C. N. Smith, The designation of chemicals in terms of the 
responses they elicit from insects, Journal of Economic Entomology, 53, 134, 1960. 

9. L. Barton Browne, Host-related responses and their suppression: Some behavioral considerations, in 
Chemical Control of Insect Behavior: Theory and Application, H. H. Shorey and J. J. McKelvey, Jr., 
(Eds.), New York: John Wiley, 1977, p. 117. 

10. M. R. Berenbaum, The chemistry of defense: Theory and practice, Proceedings of the National 
Academy of Sciences USA, 92, 2, 1995. 

11. V. Erspamer, Bioactive secretions of the amphibian integument, in Amphibian Biology, 
H. Heatwole and G. T. Barthalmus (Eds.), Vol. 1, Chipping Norton, NSW: Surrey Beatty & 
Sons, 1994, p. 178. 

12. G. Fritz, A. S. Rand, and C. W. de Pamphilis, The aposematically colored frog, Dendrobates 
pumilio, is distasteful to the large, predatory ant, Paraponera clavata, Biotropica, 13, 158, 1981. 

13. W. A. Szelistowski, Unpalatability of the poison arrow frog Dendrobates pumilio to the ctenid 
spider Cupiennius coccineus, Biotropica, 17, 345, 1985. 

14. E. D. Brodie, Jr., D. R. Formanowicz, Jr., and E. D. Brodie III, The development of noxiousness 
of Bufo americanus tadpoles to aquatic insect predators, Herpetologica, 34, 302, 1978. 

15. D. R. Formanowicz, Jr., and E. D. Brodie, Jr.. Relative palatabilities of members of a larval 
amphibian community, Copeia, 1982(1), 91, 1982. 

16. J. A. Peterson and A. R. Blaustein, Relative palatabilities of anuran larvae to natural aquatic insect 
predators, Copeia, 1992(2), 577, 1992. 

17. E. D. Brodie, Jr., and D. R. Formanowicz, Jr., Larvae of the predaceous diving beetle Dytiscus 
verticalis acquire an avoidance response to skin secretions of the newt Notophthalmus viridescens, 
Herpetologica, 37, 172, 1981. 

18. E. D. Brodie, Jr., R. T. Nowak, and W. R. Harvey, The effectiveness of antipredator secretions and 
behavior of selected salamanders against shrews, Copeia, 1979(2), 270, 1979. 

19. F. H. Pough, Leech-repellent property of eastern red-spotted newts, Notophthalmus viridescens. 
Science, 174, 1144, 1971. 

20. K. S. Hileman, E. D. Brodie, Jr., and D. R. Formanowicz, Jr., Avoidance of unpalatable prey by 
predaceous diving beetle larvae: The role of hunger level and experience (Coleoptera: Dytiscidae), 
Journal of Insect Behavior, 8, 241, 1995. 

21. M.-O. Rodel and U. Braun. Associations between anurans and ants in a West African savanna 
(Anura: Microhylidae, Hyperoliidae, and Hymenoptera: Formicidae), Biotropica, 31, 178, 1999. 

22. A. Dejean and J.-L. Amiet, Un cas de myrmecophilie inattendu: La cohabitation de l'anoure 
Kassina senegalensis avec la fourmi Megaponera foetans, Alytes, 10, 31, 1992. 

23. A. Schliiter and J. Regos, The tadpole of Lithodytes lineatus —with note on the frogs [sic] 
resistance to leaf-cutting ants (Amphibia: Leptodactylidae), Stuttgarter Beitrage zur Naturkunde, 
Series A, 536, 1, 1996. 

24. C. R. Williams, J. F. Wallman, and M. J. Tyler, Toxicity of green tree frog (Litoria caerula ) skin 
secretion to the blowflies Calliphora stygia (Fabricus) and Lucilia cuprina (Wiedemann) (Diptera: 
Calliphoridae), Australian Journal of Entomology, 37, 85, 1998. 

25. B. P. C. Smith, C. R. Williams, M. J. Tyler, andB. D. Williams, A survey of frog odorous secretions, their 
possible functions and phylogenetic significance. Applied Herpetology, 2, 47, 2004. 

26. C. R. Williams, B. P. C. Smith, S. M. Best, and M. J. Tyler, Mosquito repellents in frog skin, Biology 
Letters, 2, 242, 2006. 

27. B. P. C. Smith, Y. Hayasaka. M. J. Tyler, and B. D. Williams, B-Caryophyllene in the skin secretion of 
the Australian green tree frog, Litoria caerulea: An investigation of dietary sources, Australian Journal 
of Zoology, 52, 521, 2004. 

28. B. P. Smith, M. J. Tyler, C. R. Williams, C. Alcaraz Zini, J. Pawliszyn, B. D. Williams, Y. Hayasaka, and 
J.M. Griffiths, First demonstration of the function of frog odour as a defensive mechanism, Unpublished 
manuscript. 

29. J. F. Watkins II, F. R. Gehlbach, and R. S. Baldridge, Ability of the blind snake, Leptotyphlops dulcis, to 
follow pheromone trails of army ants, Neivamyrmex nigrescens and N. opacithorax. The Southwestern 
Naturalist, 12, 455, 1967. 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


69 


30. F. R. Gehlbach, J. F. Watkins II, and H. W. Reno, Blind snake defensive behavior elicited by ant attacks, 
BioScience, 18, 784, 1968. 

31. M. S. Blum, J. B. Byrd, J. R. Travis, J. F. Watkins II, and F. R. Gehlbach, Chemistry of the cloacal sac 
secretion of the blind snake Leptotyphlops dulcis , Comparative Biochemistry and Physiology, 38B, 103, 
1971. 

32. Holm, P, Personal communication, 2002. 

33. H. B. Cott, The edibility of birds: Illustrated by five years’ experiments and observations 
(1941-1946) on the food preferences of the hornet, cat and man; and considered with special 
reference to the theories of adaptive coloration. Proceedings of the Zoological Society of London, 
116, 371, 1947. 

34. J. P. Dumbacher and S. Pruett-Jones, Avian chemical defense, in Current Ornithology, V. Nolan, 
Jr. and E. D. Ketterson (Eds.), Vol. 13, New York: Plenum, 1996, p. 137. 

35. J.-M. Thiollay, Foraging, home range use and social behaviour of a group-living rainforest raptor, 
the red-throated caracara Daptrius americanus. Ibis, 133, 382, 1991. 

36. J.-M. Thiollay, Personal communication, 1996. 

37. G. H. Lowery, Jr. and W. W. Dalquest, Birds from the state of Veracruz, Mexico, University of 
Kansas Publications, Museum of Natural History, 3, 531, 1951. 

38. P. J. Weldon and J. H. Rappole, A survey of birds odorous or unpalatable to humans: Possible 
indications of chemical defense. Journal of Chemical Ecology, 23, 2609, 1997. 

39. J. P. Dumbacher, B. M. Beehler, T. F. Spande, H. M. Garraffo, and J. W. Daly, Homo- 
batrachotoxin in the genus Pitohui: Chemical defense in birds?, Science, 258, 799, 1992. 

40. J. P. Dumbacher, T. F. Spande, and J. W. Daly, Batrachotoxin alkaloids from passerine birds: A 
second toxic bird genus ( Ifrita kowaldi ) from New Guinea, Proceedings of the National Academy 
of Science USA, 97, 12970, 2000. 

41. J. P. Dumbacher, A. Wako, S. R. Derrickson, A. Samuelson, T. F. Spande, and J. W. Daly, 
Melyrid beetles ( Chore sine ): A putative source for batrachotoxin alkaloids found in poison-dart 
frogs and toxic passerine birds. Proceedings of the National Academy of Science USA, 101, 
15857, 2004. 

42. P. Escalante, V. Emanuel, and J. W. Daly, Alkaloids in extracts of feathers of the red warbler. 
Journal of Ornithology, 135, 410, 1994. 

43. B. O. Poulsen, Poison in Pitohui birds: Against predators or ectoparasites?. Emu, 94, 128, 1994. 

44. K. N. Mouritsen and J. Madsen, Toxic birds: Defence against parasites?, Oikos, 69, 357, 1994. 

45. M. Pruett-Jones and S. Pruett-Jones, Analysis and ecological correlates of tick burdens in a New 
Guinea avifauna, in Bird-Parasite Interactions, J. E. Loye and M. Zuk (Eds.), Oxford: Oxford 
University Press, 1991, p. 154. 

46. J. P. Dumbacher, Evolution of toxicity in pitohuis, I. Effects of homobatrachotoxin on chewing 
lice (order Phthiraptera), Auk, 116, 957, 1999. 

47. E. H. Burtt, Jr., Think small, Auk, 116, 878, 1999. 

48. J. S. Beadell, E. Gering, J. Austin, J. P. Dumbacher, M. A. Pierce, T. K. Pratt, C. T. Atkinson, 

and R. C. Fleischer, Prevalence and differential host-specificity of two avian blood parasite 
genera in the Australo-Papuan region. Molecular Ecology, 13, 3829, 2004. 

49. H. L. Jones, Hematozoa from montane forest birds in Papua New Guinea, Journal of Wildlife 
Diseases, 21, 7, 1985. 

50. H. D. Douglas III, J. E. Co, T. H. Jones, and W. E. Connor, Heteropteran chemical repellents 
identified in the citrus odor of a seabird (crested auklet: Aethia cristatella ): Evolutionary 
convergence in chemical ecology, Naturwissenschaften, 88, 330, 2001. 

51. H. D. Douglas III, J. E. Co, T. H. Jones, and W. E. Connor, Interspecific differences in Aethia 
spp. auklet odorants and evidence for chemical defense against ectoparasites, Journal of Chemical 
Ecology, 30, 1921, 2004. 

52. H. D. Douglas III, J. E. Co, T. H. Jones, W. E. Connor, and J. F. Day, Chemical odorant of 
colonial seabird repels mosquitoes, Journal of Medical Entomology, 42, 647, 2005. 

53. H. D. Douglas III, J. R. Malenke, and D. H. Clayton, Is the citrus-like plumage odorant of crested auklets 
(Aethia cristatella) a defense against lice?. Journal of Ornithology, 146, 111, 2005. 


2006 by Taylor & Francis Group, LLC 



70 


Insect Repellents: Principles , Methods, and Uses 


54. B. R. Moyer, A. N. Rock, and D. H. Clayton, Experimental test of the importance of preen oil in rock 
doves ( Columba livid). Auk, 120, 490, 2003. 

55. R. Galun, Responses of blood-sucking arthropods to vertebrate hosts, in Chemical Control of Insect 
Behavior: Theory and Application, H. H. Shorey and J. J. McKelvey, Jr., (Eds.), New York: Wiley, 1977, 
p. 103. 

56. N. K. Gikonyo, A. Hassanali, P. G. N. Njagi, and R. K. Saini, Behaviour of Glossina morsitans morsitans 
Westwood (Diptera: Glossinidae) on waterbuck Kobus defassa Ruppel and feeding membranes smeared 
with waterbuck sebum indicates the presence of allomones, Acta Tropica, 77, 295, 2000. 

57. T. A. M. Nash, Africa’s Bane: The Tsetse Fly, London: Collins, 1969. 

58. G. A. Vale, D. R. Hall, and A. J. E. Gough, The olfactory responses of tsetse flies, Glossina spp. 
(Diptera: Glossinidae), to phenols and urine in the field. Bulletin of Entomological Research, 78, 
293, 1988. 

59. S. J. Torr, T. N. C. Mangwiro, and D. R. Hall, Responses of Glossina pallidipes (Diptera: Glossinidae) 
to synthetic repellents in the field. Bulletin of Entomological Research, 86, 609, 1996. 

60. L. C. Madubunyi, A. Hassanali, W. Ouma, D. Nyarango, and J. Kabii, Chemoecological role of 
mammalian urine in host location by tsetse, Glossina spp. (Diptera; Glossinidae), Journal of 
Chemical Ecology, 22, 1187, 1996. 

61. G. A. Vale, Field studies of the responses of tsetse flies (Glossinidae) and other Diptera to carbon 
dioxide, acetone and other chemicals. Bulletin of Entomological Research, 70, 563, 1980. 

62. W. F. Wood, A. Walsh, J. Seyjagat, and P. J. Weldon, Volatile compounds in shoulder gland 
secretions of male flying foxes, genus Pteropus (Pteropodidae, Chiroptera), Zeitschrift fur 
Naturforschung C, 60, 779, 2005. 

63. N. K. Gikonyo, A. Hassanali, P. G. N. Njagi, P. M. Gitu, and J. O. Midiwo, Odor composition of 
preferred (buffalo and ox) and nonpreferred (waterbuck) hosts of some savanna tsetse flies. 
Journal of Chemical Ecology, 28, 969, 2002. 

64. N. K. Gikonyo, A. Hassanali, P. G. N. Njagi, and R. K. Saini, Responses of Glossina morsitans 
morsitans to blends of electroantennographically active compounds in the odors of its preferred 
(buffalo and ox) and nonpreferred (waterbuck) hosts. Journal of Chemical Ecology, 29, 2331, 
2003. 

65. M. Bett, R. K. Saini, and A. Hassanali, In preparation. 

66. P. J. Weldon, Unpublished. 

67. P. J. Weldon, J. R. Aldrich, J. A. Klun, J. E. Oliver, and M. Debboun, Benzoquinones from 
millipedes deter mosquitoes and elicit self-anointing in capuchin monkeys ( Cebus spp.), 
Naturwissenschaften, 90, 301, 2003. 

68. J. E. Oliver, P. J. Weldon, K. S. Petersen, W. F. Schmidt, and M. A. Debboun, Mosquito- 
deterring acid from pelage of gaur. Bos gaurus, Abstracts of Papers, 225th National Meeting of 
the American Chemical Society, New Orleans, LA, March 23-21. Washington, DC: American 
Chemical Society, 2003. 

69. H. Ishii, S. Krane, Y. Itagaki, N. Berova, K. Nakanishi, and P. J. Weldon, Absolute configuration 
of a hydroxyfuranoid acid from the pelage of the genus Bos, 18-(65,9R,10R)-bovidic acid. 
Journal of Natural Products, 67, 1426, 2004. 

70. A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and R. K. Gupta, 3D Pharmacophore model 
for insect repellent activity and discovery of new repellent candidates, QSAR & Combinatorial 
Science, 24, 593, 2005. 

71. P. A. Evans, W. J. Andrews, and P. J. Weldon, Unpublished. 

72. F. Q. Alali, X.-X. Liu, and J. L. McLaughlin, Annonaceous acetogenins: Recent progress. Journal 
of Natural Products, 62, 504, 1999. 

73. C. Costantini, G. Gibson, N'F. Sagnon, A. della Torre, J. Brady, and M. Coluzzi, Mosquito 
responses to carbon dioxide in a West African Sudan savanna village, Medical & Veterinary 
Entomology, 10, 220, 1996. 

74. C. Costantini, N’F. Sagnon, A. della Torre, M. Diallo, J. Brady, G. Gibson, and M. Coluzzi, 
Odor-mediated host preferences of West African mosquitoes, with particular reference to malaria 
vectors, American Journal of Tropical Medicine and Hygiene, 58, 56, 1998. 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


71 


75. M. T. Gillies, Experiments on host selection in the Anopheles gambiae complex. Annals of 
Tropical Medicine and Parsitology, 61, 68, 1967. 

76. T. Dekker and W. Takken, Differential responses of mosquito sibling species Anopheles 
arabiensis and An. quadriannulatus to carbon dioxide, a man or a caff. Medical & Veterinary 
Entomology , 12, 136, 1998. 

77. T. Dekker, W. Takken, and M. A. H. Braks, Innate preference for host-odor blends modulates 
degree of anthropophagy of Anopheles gambiae sensu lato (Diptera: Culicidae), Journal of 
Medical Entomology , 38, 868, 2001. 

78. H. V. Pates, W. Takken, C. F. Curtis, P. W. Huisman, O. Akinpelu, and G. S. Gill, Unexpected 
anthropophagic behaviour in Anopheles quadriannulatus, Medical & Veterinary Entomology, 15, 
293, 2001. 

79. H. V. Pates, W. Takken, K. Stuke, and C. F. Curtis, Differential behaviour of Anopheles gambiae 
sensu stricto (Diptera: Culicidae) to human and cow odours in the laboratory, Bulletin of 
Entomological Research, 91, 289, 2001. 

80. K. M. V. Jensen, J. B. Jespersen, M. A. Birkett, J. A. Pickett, G. Thomas, L. J. Wadhams, and 
C. M. Woodcock, Variation in the load of the horn fly, Haematobia irritans, in cattle herds is 
determined by the presence or absence of individual heifers. Medical & Veterinary Entomology, 
18, 275, 2004. 

81. G. Thomas, cited in Pickett and Woodcock, 1996 (Ref.83). 

82. M. A. Birkett, N. Agelopoulos, K. M. V. Jensen, J. B. Jespersen. J. A. Pickett, H. J. Prijs, 
G. Thomas, J. J. Trapman, L. J. Wadhams, and C. M. Woodcock, The role of volatile 
semiochemicals in mediating host location and selection by nuisance and disease-transmitting 
cattle flies. Medical & Veterinary Entomology, 18, 313, 2004. 

83. J. A. Pickett and C. M. Woodcock, The role of mosquito olfaction in oviposition site location 
and in the avoidance of unsuitable hosts, in Olfaction in Mosquito-Host Interactions, G. R. Bock 
and G. Cardew (Eds.), New York: Wiley, Ciba Foundation Symposium 200, 1996, p. 109. 

84. Y. S. Pan, Breed and seasonal differences in quantities of lipids on skin surface and hair in cattle. 
Journal of Agricultural Science, 75, 41, 1970. 

85. J. C. Bonsma, Breeding tick-repellent cattle, in Tick Biology and Control: Proceedings of an 
International Conference, G. B. Whitehead and J. D. Gibson (Eds.), Tick Research Unit, Rhodes 
University: Grahamstown, 1981, p. 67. 

86. W. Wanzala, N. F. K. Sika, S. Gule, and A. Hassanali, Attractive and repellent host odours guide 
ticks to their respective feeding sites, Chemoecology, 14, 229, 2004. 

87. N. F. K. Sika, Behavioural responses of Rhipicephalus appendiculatus Neumann 1901 and 
Rhipicephalus evertsi Neumann 1897 (Acari: Ixodidae) to host and non-host semiochemicals. 
Ph.D. dissertation, Kenyatta University, Kenya, 1996. 

88. J. F. Carroll, J. A. Klun, and E. T. Schmidtmann, Evidence for kairomonal influence on selection 
of host-ambushing sites by adult Ixodes scapularis (Acari: Ixodidae), Journal of Medical 
Entomology, 32, 119, 1995. 

89. J. F. Carroll, Responses of three species of adult ticks (Acari: Ixodidae) to chemicals in the coats 
of principal and minor hosts, Journal of Medical Entomology, 36, 238, 1999. 

90. J. F. Carroll, Interdigital gland substances of white-tailed deer and the response of host-seeking 
ticks (Acari: Ixodidae), Journal of Medical Entomology, 38, 114, 2001. 

91. J. F. Carroll, Notes on responses of blacklegged ticks (Acari: Ixodidae) to host urine, Journal of 
Medical Entomology, 36, 212, 1999. 

92. J. F. Carroll, Responses of adult Ixodes scapularis (Acari: Ixodidae) to urine produced by 
white-tailed deer of various reproductive conditions. Journal of Medical Entomology, 37, 472, 
2000 . 

93. G. A. Vale, Mobile attractants for tsetse flies, Arnoldia, 4, 1, 1969. 

94. G. A. Vale, New field methods for studying the responses of tsetse flies (Diptera, Glossinidae) to 
hosts. Bulletin of Entomological Research, 64, 199, 1974. 

95. G. A. Vale, The responses of tsetse flies (Diptera, Glossinidae) to mobile and stationary hosts. 
Bulletin of Entomological Research, 64, 545, 1974. 


2006 by Taylor & Francis Group, LLC 



72 


Insect Repellents: Principles, Methods, and Uses 


96. G. A. Vale. Feeding responses of tsetse flies (Diptera: Glossinidae) to stationary hosts, Bulletin of 
Entomological Research, 67, 635, 1977. 

97. J. W. Hargrove, The effect of human presence on the behaviour of tsetse ( Glossina spp.) (Diptera, 
Glossinidae) near a stationary ox, Bulletin of Entomological Research, 66, 173, 1976. 

98. G. A. Vale, Field responses of tsetse flies (Diptera: Glossinidae) to odours of men, lactic acid and 
carbon dioxide. Bulletin of Entomological Research, 69, 459, 1979. 

99. B. M. Steib, M. Geier, and J. Boeckh, The effect of lactic acid on odour-related host preference of 
yellow fever mosquitoes. Chemical Senses, 26, 523, 2001. 

100. A. W. A. Brown, D. S. Sarkaria, and R. P. Thompson, Studies on the responses of the female Aedes 
mosquito, Part I. The search for attractant vapours, Bulletin of Entomological Research, 42, 105, 1951. 

101. R. P. Thompson and A. W. A. Brown, The attractiveness of human sweat to mosquitoes and the role of 
carbon dioxide, Mosquito News, 15, 80, 1955. 

102. M. R. Smart and A. W. A. Brown, Studies on the responses of the female Aedes mosquito. Part VII. The 
effect of skin temperature, hue and moisture on the attractiveness of the human hand. Bulletin of 
Entomological Research, 47, 89, 1956. 

103. H. I. Maibach, W. A. Skinner, W. G. Strauss, and A. A. Khan, Factors that attract and repel mosquitoes in 
human skin, Journal of the American Medical Association, 196, 173, 1966. 

104. W. Muller, Die Distanz- und Kontakt-Orientierung der Stechmiicken ( Aedes aegypti) (Wirtsfindung, 
Stechverhalten und Blutmahlzeit), Zeitschrift fur Vergleichende Physiologie, 58, 241, 1968. 

105. W. A. Skinner, H. Tong, H. Maibach, A. A. Khan, and T. Pearson, Repellency of skin-surface lipids of 
humans to mosquitoes, Science, 149, 305, 1965. 

106. W. A. Skinner, H. C. Tong, T. R. Pearson, and H. I. Maibach. Repellency of human skin-surface lipid 
hydrocarbons to the yellow-fever mosquito, Journal of Economic Entomology, 60, 927, 1967. 

107. W. A. Skinner, H. C. Tong, H. I. Maibach, and D. Skidmore, Human skin-surface lipid fatty acids— 
mosquito repellents, Experientia, 26, 728, 1970. 

108. W. A. Skinner, H. Tong, H. Johnson, H. Maibach, and D. Skidmore, Human sweat components— 
attractancy and repellency to mosquitoes, Experientia, 24, 679, 1968. 

109. O. J. Bosch, M. Geier, and J. Boeckh, Contribution of fatty acids to olfactory host finding of female Aedes 
aegypti, Chemical Senses, 25, 323, 2000. 

110. B. G. J. Knols, J. J. A. van Loon, A. Cork, R. D. Robinson, W. Adam, J. Meijerink, R. De Jong, and 
W. Takken, Behavioural and electrophysiological responses of the female malaria mosquito Anopheles 
gambiae (Diptera: Culicidae) to Limburger cheese volatiles, Bulletin of Entomological Research, 87, 
151, 1997. 

111. R. C. Smallegange, Y. T. Qiu, J. J. van Loon, and W. Takken, Syngergism between ammonia, lactic acid 
and carboxylic acids as kairomones in the host-seeking behaviour of the malaria mosquito Anopheles 
gambiae sensu stricto (Diptera: Culicidae), Chemical Senses, 30, 145, 2005. 

112. Y. Shirai. K. Kamimura, T. Seki, and M. Morohashi, L-lactic acid as a mosquito (Diptera: Culicidae) 
repellent on human and mouse skin. Journal of Medical Entomology, 38, 51, 2001. 

113. C. N. Smith, N. Smith, H. K. Gouck, D. E. Weidhaas, I. H. Gilbert, M. S. Mayer, B. J. Smittle, and 
A. Hofbauer, L-lactic acid as a factor in the attraction of Aedes aegypti (Diptera: Culicidae) to human 
hosts, Annals of the Entomological Society of America, 63, 760, 1970. 

114. C. Costantini, M. A. Birkett, G. Gibson, J. Ziesmann, N. F. Sagnon, H. A. Mohammed, M. Coluzzi, and 
J. A. Pickett, Electroantennogram and behavioural responses of the malaria vector Anopheles gambiae 
to human-specific sweat components, Medical & Veterinary Entomology, 15, 259, 2001. 

115. W. R. Mukabana, W. Takken, G. F. Killeen, and B. G. J. Knols, Allomonal effect of breath contributes to 
differential attractiveness of humans to the African malaria vector Anopheles gambiae. Malaria 
Journal, 3, 1, 2004. 

116. K. E. McKenzie, The occurrence of preferential host selection by Aedes aegypti (Linnaeus 1762) 
(Diptera: Culicidae), M.S. thesis. University of Florida, Gainesville, 2000. 

117. U. R. Bernier, D. L. Kline, C. E. Schreck, R. A. Yost, and D. R. Barnard, Chemical analysis of human 
skin emanations: Comparison of volatiles front humans that differ in attraction of Aedes aegypti 
(Diptera: Culicidae), Journal of the American Mosquito Control Association, 18,186, 2002. 

118. J. Kingdon, East African Mammals: Vol. IIIA, Carnivores, Chicago: University of Chicago Press, 1977. 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


73 


119. A. M. Young. A human sweat-mediated defense against multiple attacks by the wasp Polybia diguetana 
in northeastern Costa Rica, Biotropica, 10, 73, 1978. 

120. L. E. Sedgwick, Magpies “anting”. Western Australian Naturalist, 1, 21, 1947. 

121. A. R. Laskey. Bronzed grackle anointing plumage with orange-skin. The Wilson Bulletin, 60. 244, 
1948. 

122. A. C. Govan, “Anting”—outside and inside the windowpane. Nature Magazine, January, 41, 1954. 

123. L. M. Whitaker, A resume of anting, with particular reference to a captive orchard oriole, The Wilson 
Bulletin, 69, 195, 1957. 

124. A. H. Chisholm, The history of anting. Emu, 59, 101, 1959. 

125. J. M. Goodman, Aves incendiaria. The Wilson Bulletin, 72, 400, 1960. 

126. L. Kelso and M. M. Nice, A Russian contribution to anting and feather mites. The Wilson Bulletin, 
75, 23, 1963. 

127. K. E. L. Simmons, Anting and the problem of self-stimulation, Journal of the Zoology (London), 149, 
145, 1966. 

128. E. F. Potter and D. C. Hauser, Relationship of anting and sunbathing to molting in wild birds. Auk, 
91, 537, 1974. 

129. J. V. Dennis, Commentary on grackles anting with marigold blossoms. Blue Jay, 43, 175, 1985. 

130. D. H. Clayton and J. G. Vernon, Common grackle anting with lime fruit and its effect on 
ectoparasites. Auk, 110. 951, 1993. 

131. H. Sick, Birds in Brazil, Princeton. NJ: Princeton University Press, 1993. 

132. K. C. Parkes. P. J. Weldon, and R. L. Hoffman, Polydesmidan millipede used in self-anointing by a 
strong-billed woodcreeper ( Xiphocolaptes promeropirhyncus ) from Belize, Ornitology Neotropical, 
14, 285, 2003. 

133. E. A. Vander Werf, ‘Elepaio “anting” with a garlic snail and a Schinus fruit, Journal of Field 
Ornithology, 76, 134, 2005. 

134. E. Gould and J. F. Eisenberg, Notes on the biology of the Tenrecidae, Journal of Mammalogy, 47, 
660, 1966. 

135. E. D. Brodie, Jr., Hedgehogs use toad venom in their own defence. Nature, 268, 627, 1977. 

136. A. Nolte, Beobachtungen iiber das Instinktverhalten von Kapuzineraffen (Cebus apella L.) in der 
Gefangenschaft, Behaviour, 13, 189, 1958. 

137. Oppenheimer, J. R. Behavior and ecology of the white-faced monkey, Cebus capucinus, on Barro 
Colorado Island, C.Z., Ph.D. dissertation, University of Illinois at Urbana-Champaign, 1968. 

138. J. Longino. True anting by the capuchin, Cebus capucinus, Primates, 25, 243, 1984. 

139. D. J. Overdorff. Similarities, differences, and seasonal patterns in the diets of Eulemur rubriventer 
and Eulemur fulvus rufus in the Ranomafana National Park, Madagascar, International Journal of 
Primatology, 14, 721, 1993. 

140. X. Valderrama, J. G. Robinson, A. B. Attygalle, and T. Eisner, Seasonal anointment with 
millipedes in a wild primate: A chemical defense against insects?. Journal of Chemical Ecology, 
26, 2781, 2000. 

141. M. DeJoseph, R. S. Taylor, M. Baker, and M. Aregullin, Fur-rubbing behavior of capuchin monkeys. 
Journal of the American Academy of Dermatology, 46, 924, 2002. 

142. M. Zito, S. Evans, and P. J. Weldon, Owl monkeys (Aotus spp.) self-anoint with plants and 
millipedes. Folia Primatologica, 74, 159, 2003. 

143. M. P. Verderane, T. Falotico, B. D. Resende, M. B. Labruna, P. Izar, and E. Ottoni. Anting in a 
semifree-ranging group of Cebus apella, International Journal of Primatology, in press. 

144. Z. Xu, D. M. Stoddart, H. Ding, and J. Zhang, Self-anointing behavior in the rice-field rat, Rattus 
rattoides. Journal of Mammalogy, 76, 1238, 1995. 

145. E. Zimen, The Wolf: A Species in Danger, New York: Delacourte, 1981. 

146. I. Reiger, Scent rubbing in carnivores. Carnivore, 2, 17, 1979. 

147. A. O. Tucker and S. S. Tucker, Catnip and the catnip response. Economic Botany, 42, 214, 
1988. 

148. P. J. Weldon and D. B. Fagre, Responses by canids to scent gland secretions of the western 
diamondback rattlesnake ( Crotalus atrox ), Journal of Chemical Ecology, 15, 1589, 1989. 


2006 by Taylor & Francis Group, LLC 



74 


Insect Repellents: Principles, Methods, and Uses 


149. M. E. Gompper and A. M. Hoylman, Grooming with Trattinnickia resin: Possible pharma¬ 
ceutical plant use by coatis in Panama, Journal of Tropical Ecology, 9, 533, 1993. 

150. S. Sigstedt, cited in Clayton and Wolfe, 1993 (Ref.3). 

151. L. M. Gosling and H. V. McKay, Scent-rubbing and status signalling by male mammals, 
Chemoecology, 1, 92, 1990. 

152. P. R. Ehrlich, D. S. Dobkin, and D. Wheye, The adaptive significance of anting. Auk, 103, 835, 1986. 

153. W. Eichler, Die Biologie der Federlinge, Journal fur Ornithologie, 84, 471, 1936. 

154. S. Wilson and N. Hillgarth, cited in Clayton and Wolfe, 1993 (Ref.3). 

155. A. Bennett and D. Clayton, cited in Clayton and Wolfe, 1993 (Ref.3). 

156. A. T. D. Bennett, N. Elillgarth and J. Vernon cited in Judson and Bennett, 1992 (Ref. 157). 

157. O. P. Judson and A. T. D. Bennett, Anting as food preparation: Formic acid is worse on an empty 
stomach. Behavioral Ecology and Sociobiology, 31, 437, 1992. 

158. J. F. Carroll, M. Kramer, P. J. Weldon, and R. G. Robbins, Anointing chemicals and ectoparasites: 
Effects of benzoquinones from millipedes on the lone star tick, Amblyomma americanum. Journal of 
Chemical Ecology, 31, 63, 2005. 

159. G. Bisignano and A. Saija, The biological activity of citrus oils, in Citrus: The Genus Citrus, G. Dugo 
and A. Di Giacomo (Eds.), New York: Taylor & Francis, 2002, p. 602. 

160. Y. Trongtokit, Y. Rongsriyam, N. Komalamisra, and C. Apiwathnasom, Comparative repellency of 38 
essential oils against mosquito bites. Phytotherapy Research, 19, 303, 2005. 

161. M. A. Oshaghi, R. Ghalandari, H. Vatandoost, M. Shayeghi, M. Kamali-Nejad, H. Tourabi-Khaledi, 
M. Abolhassani, and M. Flashemzadeh, Repellent effect of extracts and essential oils of Citrus Union 
(Rutaceae) and Melissa officinalis (Labiatae) against main malaria vector. Anopheles stephensi 
(Diptera: Culicidae), Iranian Journal of Public Health, 32, 47, 2003. 

162. A. A. Shalaby, K. A. Allam, A. A. Mostafa, and S. M. Fahmy, Insecticidal properties of citrus oils against 
Culex pipiens and Musca domestica. Journal of the Egyptian Society of Parasitology, 28, 595, 1998. 

163. P. J. Weldon, Unpublished. 

164. G. Dugo, A. Cotroneo, A. Verzera, and I. Bomaccorsi, Composition of the volatile fraction of cold- 
pressed citrus peel oils, in Citrus: The Genus Citrus, G. Dugo and A. Di Giacomo (Eds.), New York: 
Taylor & Francis, 2002, p. 201. 

165. D. M. Norris, Repellents, in CRC Handbook of Natural Pesticides, Vol. VI, Insect Attractants 
and Repellents, E. D. Morgan and N. B. Mandava (Eds.), Boca Raton: CRC Press, 1996, p. 135. 

166. S. Dev and O. Koul, Insecticides of Natural Origin, Amsterdam: Harwood, 1997. 

167. W. F. Hink and B. J. Fee, Toxicity of D-limonene, the major component of citrus peel oil, to all life 
stages of the cat flea, Ctenocephalides felis (Siphonaptera: Pulicidae), Journal of Medical 
Entomology, 23, 400, 1986. 

168. W. F. Hink, T. A. Fiberati, and M. G. Collart, Toxicity of linalool to life stages of the cat flea, 
Ctenocephalides felis (Siphonaptera: Pulicidae), and its efficacy in carpet and on animals. Journal 
of Medical Entomology, 25, 1, 1988. 

169. K. Y. Mumcuoglu, R. Galun, U. Bach, J. Miller, and S. Magdassi, Repellency of essential oils and 
their components to the human body louse, Pediculus humanus humanus, Entomologia Experi- 
mentalis et Applicata, 78, 309, 1996. 

170. J. F. Carroll, Feeding deterrence of northern fowl mites (Acari: Macronyssidae) by some naturally 
occurring plant substances. Pesticide Science, 41, 203, 1994. 

171. T. Eisner, Catnip: Its raison d’etre. Science, 146, 1318, 1964. 

172. C. J. Petersen, Insect repellents of natural origin: Catnip and osage orange, Ph.D. dissertation, Iowa 
State University, Ames, 2001. 

173. U. R. Bernier, K. D. Furman, D. F. Kline, S. A. Allan, and D. R. Barnard, Comparison of contact 
and spatial repellency of catnip oil and (V../V-diethyl-3-methylbenzamide (deet) against mosquitoes. 
Journal of Medical Entomology, 42, 306, 2005. 

174. K. R. Chauhan, J. A. Klun, M. Debboun, and M. Kramer, Feeding deterrent effects of catnip oil 
components compared with two synthetic amides against Aedes aegypti. Journal of Medical 
Entomology, 42, 643, 2005. 

175. W. Wickler, Mimicry in Plants and Animals, New York: McGraw-Hill, 1968. 


© 2006 by Taylor & Francis Group, LLC 



Vertebrate Chemical Defense: Secreted and Topically Acquired Deterrents of Arthropods 


75 


176. C. M. Passreiter, Y. Akhtar, and M. B. Isman, Insecticidal activity of the essential oil of 
Ligusticum mutellina roots, Zeitschrift fiir Naturforschung C, 60, 411, 2005. 

177. P. T. Fauth, D. G. Krementz, and J. E. Hines, Ectoparasitism and the role of green nesting material 
in the European starling, Oecologia, 88, 22, 1991. 

178. H. Gwinner, M. Oltrogge, L. Trost, and U. Nienaber, Green plants in starling nests: Effects on 
nestlings. Animal Behaviour, 59, 301, 2000. 

179. P. H. Wimberger, The use of green plant material in bird nests to avoid ectoparasites. Auk, 101, 
615, 1984. 

180. L. Clark and J. R. Mason, Use of nest material as insecticidal and anti-pathogenic agents by the 
European starling, Oecologia, 67. 169, 1985. 

181. S. Sengupta, Adaptive significance of the use of margosa leaves in nests of house sparrows Passer 
domesticus. Emu, 81, 114. 1981. 

182. L. Clark and J. R. Mason. Effect of biologically active plants used as nest material and the derived 
benefit to starling nestlings, Oecologia, 77, 174, 1988. 

183. J. R. Mason and L. Clark, Chemoreception and the selection of green plants as nest fumigants by 
starlings, in Chemical Signals in Vertebrates 4: Ecology, Evolution, and Comparative Biology, 
D. Duvall, D. Miiller-Schwarze, and R. M. Silverstein (Eds.), New York: Plenum Press, 1986, p. 369. 

184. C. Petit, M. Hossaert-McKey, P. Perret, J. Blondel, and M. M. Lambrechts, Blue tits use selected 
plants and olfaction to maintain an aromatic environment for nestlings, Ecology Letters, 5, 585, 
2002 . 

185. L. Clark and C. A. Smeraski, Seasonal shifts in odor acuity by starlings, Journal of Experimental 
Zoology, 255, 22, 1990. 

186. L. Lafurna, M. M. Lambrechts, and M. Raymond, Aromatic plants in bird nests as a protection 
against blood-sucking flying insects?. Behavioural Processes, 56, 113, 2001. 

187. R. B. Hemmes, A. Alvarado, and B. L. Hart, Use of California bay foliage by wood rats for possible 
fumigation of nest-borne ectoparasites. Behavioral Ecology, 13, 381, 2002. 

188. D. J. Brightsmith, Use of arboreal termitaria by nesting birds in the Peruvian Amazon, Condor, 102, 
529, 2000. 

189. D. H. Janzen, Birds and the antXacacia interaction in Central America, with notes on birds and other 
myrmecophytes. Condor, 71, 240, 1969. 

190. N. Nicolaides, Skin lipids: Their biochemical uniqueness, Science, 186, 19, 1974. 

191. P. J. Weldon, Thin-layer chromatography of the skin secretions of vertebrates, in Practical Thin-layer 
Chromatography: A Multidisciplinary Approach, B. Fried and J. Sherma (Eds.). Boca Raton: CRC 
Press, 1996, p. 105. 

192. A. M. Kligman, The uses of sebum?, in The Sebaceous Glands, W. Montagna, R. A. Ellis, and 
A. F. Silver (Eds.), New York: Pergamon Press, 1963, p. 110. 

193. W. F. Wood and P. J. Weldon, The scent of the reticulated giraffe (Giraffa Camelopardalis 
reticulata). Biochemical Systematics and Ecology, 30, 913, 2002. 

194. W. Rudolfs, Chemotropism of mosquitoes. New Jersey Agricultural Experiment Station Bulletin, 
367, 4, 1922. 

195. W. Rudolfs, Effects of chemicals upon the behavior of mosquitoes. New Jersey Agricultural 
Experiment Station Bulletin, 496, 2, 1930. 

196. W. F. Wood, M. G. Leahy, R. Galun, G. D. Prestwich, J. Meinwald, R. E. Purnell, and R. C. Payne, 
Phenols as pheromones of ixodid ticks: A general phenomenon?, Journal of Chemical Ecology, 1, 
501, 1975. 

197. J. A. Pickett, L. J. Wadhams, and C. M. Woodcock, Insect supersense: Mate and host location by 
insects as model systems for exploiting olfactory interactions, The Biochemist, August, 8, 1998. 

198. T. Eisner and R. P. Grant, Toxicity, odor aversion, and “olfactory aposematism”. Science, 213, 476, 
1981. 


2006 by Taylor & Francis Group, LLC 



© 2006 by Taylor & Francis Group, LLC 



4 


Human Emanations and Related Natural Compounds 
That Inhibit Mosquito Host-Finding Abilities 


Ulrich R. Bernier, Daniel L. Kline, and Kenneth H. Posey 


CONTENTS 

Concepts and Terminology Used in This Chapter.78 

Attraction Antagonists and Anti-Attractants.78 

Spatial Repellents and Irritants.79 

Attraction-Inhibitors.80 

Spatial Repellency and Attraction-Inhibition Research.80 

Early History of Spatial Repellents Testing.80 

Spatial Repellency and Attraction-Inhibition of Catnip Oil.81 

Spatial Repellency and Attraction-Inhibition of Deet.81 

Attraction-Inhibition by Linalool and Related Compounds.82 

Human-Produced Compounds That Affect Host-Seeking.82 

Attraction-Inhibition by Carboxylic Acids.85 

Attraction-Inhibition by Aldehydes.85 

Attraction-Inhibition by Ketones.86 

Attraction-Inhibition by Alcohols.86 

Attraction-Inhibition by Compounds of Other Classes.87 

Identification of Host-Produced Allelochemicals.87 

Analysis of Human Emanations.88 

Merging Chemistry and Sensory Physiology.88 

Current State and Future Directions of Host Odor Research.89 

Laboratory Bioassays of Spatial Repellents and Attraction-Inhibitors.89 

Olfactometers for the Assessment of Spatial Repellents.89 

Olfactometers for the Assessment of Attraction-Inhibitors.90 

Considerations in the Experimental Design.90 

Correlating Small- and Large-Scale Laboratory Results to Field Experiments.91 

Field Tests and Use of Spatial Repellents and Attraction-Inhibitors.92 

Experimental Design of Field Tests.92 

Use of Large-Cage Experiments and Laboratory-Reared Colony Mosquitoes.92 

Experiments with Wild Mosquitoes in the Field.92 

Use of Stand-Alone Inhibitor-Delivery Technology.93 

Potential Applications of Spatial Repellents and Attraction-Inhibitors.93 

Species-Specific or Species-Exclusive Trapping.93 

77 


© 2006 by Taylor & Francis Group, LLC 




































78 


Insect Repellents: Principles, Methods, and Uses 


Local Area Host-Finding Reduction.93 

Local Control Using a Push-Pull Strategy with Attractant-Baited Surveillance Traps.93 

Use of Structure-Activity Relationships to Benefit Development of Attraction-Inhibitors.94 

References.94 


The beginning of knowledge is the discovery of something we do not understand. 


l 


Concepts and Terminology Used in This Chapter 

One mechanism by which the action of semiochemicals can be classified is based on the behavioral 
impact within or external to the species of interest. As such, one can classify a chemical as one of the 
following 2,3 : 

1. Pheromone, if it results in response between insects of the same species 

2. Kairomone, if it results in response in another species that benefits the species receiving the 
chemical cue 

3. Allomone, if it results in response in another species that benefits the species releasing the 
chemical cue 

However, the distinctions can be more specific by classification of chemical cues through the imparted 
behavioral effect: attractant; repellent; arrestant; locomotory stimulant; feeding, mating, or oviposition 
stimulant; and feeding, mating, or oviposition deterrent. 2,4 Karlson and Liischer first proposed the term 
“pheromone” to describe chemicals with instraspecific species activity. 5,6 Chemicals with interspecific 
species activity are allelochemicals. 3 Allelochemicals can be separated further into kairomones, of which 
attractants are a category of, and allomones, which are the primary focus of both this book and chapter, 
and the class that repellents are a part of. Furthermore, attraction-inhibitors, are classified by us as a 
category of repellents. Ironically, many of the attraction-inhibitors have been discovered in a search for 
kairomones used by mosquitoes to locate human hosts. Attraction-inhibitors may not repel by the 
traditional mechanisms, but they do interfere, or act as an antagonist to the normal attraction response of 
an insect to attractive odor(s). 

The proper name for the behavioral actions that are described in this chapter can be debated 
extensively and additional discussion of terminology is found in Chapter 2 by White. In this short 
prequel to the main body of our contribution on human and other compounds that interfere with mosquito 
host-finding, we put forth our rationale supporting the terms used to describe behaviors reported in 
this chapter. 

Attraction Antagonists and Anti-Attractants 

Attraction antagonist is an appropriate term to describe compounds that interrupt the blood-feeding 
process in bioassays. The term “antagonism” has been used to describe a phenomenon between two 
toxicants that is the opposite effect of synergism. 7,8 Applying this by analogy to attraction, a synergistic 
response is then a response where the combination of chemicals in a blend produces a level of attraction 
greater than the sum of attraction response levels to the single substances. Thus, antagonism describes a 
situation where the attraction to a combination is less than the sum of individual attraction levels. 


© 2006 by Taylor & Francis Group, LLC 








Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


79 


Wright et al. 9 stated that, in principle, these compounds also function as “anti-attractants” in that they 
disrupt the function of naturally occurring attractants. Wright et al. noted furthermore that repellents and 
anti-attractants should be considered as separate functional classes of compounds based on their different 
modes of action. 

It is commonly accepted that a volatile chemical attractant is a substance that produces oriented insect 
movement (positive taxis) or upwind movement (anemotaxis) toward a source by following a 
concentration gradient of gas-phase molecules distributed in plumes. 10 Therefore, a broad term is 
needed to describe compounds that prevent host finding by interference of the positive anemotaxis 
without too much reliance on characterizing the mechanism of action on the insect. Bearing this mind, 
compounds that repel using the criteria of Dethier et al. 4 would be classified as antagonists because the 
repellent substances prevent host finding by an oriented movement away from the source. However, 
compounds that cloak or hide the host from mosquitoes that would otherwise be able to locate the host for 
a blood meal would not fit the strict definition of repellents in the sense of Dethier et al. because the 
mosquitoes would not necessarily exhibit oriented movement away from the source. 4 


Spatial Repellents and Irritants 

Spatial repellency and irritancy can involve more than simple concealment of host location or attractant 
odor source. As noted above, a rigorous definition of “repellent” requires movement away from the 
source. Barton Browne later suggested that “movement away” is not necessarily a suitable criterion, and 
that “a repellent is almost always assessed in terms of its ability to inhibit the insect’s response to 
chemical attractants...” 10 This led to the proposal that a repellent is a chemical that, acting in the vapor 
phase, prevents an insect from reaching a target to which it would otherwise be attracted. However, it 
should be noted that vapor-phase activity might not be necessary to repel. If contact is made with a 
surface that contains repellent, then the mosquito chemoreceptors can detect this repellent if it is present 
at the required threshold concentration to cause repellency. Additionally, it should be apparent that 
chemical compounds have a vapor phase concentration that is dependent upon their volatilities, and 
that this concentration falls off as the distance from the source increases. Therefore, the true criterion is 
linked to the threshold level of chemoreception of (repellent) molecules by the mosquito to elicit the 
desired behavior (repellency). Obviously, this can occur in space if the mosquito has high sensitivity to 
the chemical and the chemical has a high vapor phase concentration. Similarly, repellent chemicals that 
the mosquito is less sensitive to, or that have low vapor phase concentration from low evaporative loss, 
will result in repellency closer to the surface, or perhaps even by contact. 

The use of the word “spatial” to classify repellents was defined by Gouck et al. 11 as a compound or 
agent that can produce repellency at a distance. Furthermore, spatial repellents have been described as 
repellents that inhibit the ability of mosquitoes to locate a target host. 12 Thus, topical repellents with low 
vapor pressure, such as !V,lV-diethyl-3-methylbenzamide (deet), and highly volatile spatial repellents 
(attraction-inhibitors) are repellents, even though their modes of action may be radically different. 

Another possible source of confusion arises from the term “area repellent.” Although spatial repellents 
should ideally prevent biting in a defined local area, an area repellent does not necessarily require a 
significant vapor phase spatial repellent effect. An example of this is the use of a repellent that is 
normally applied topically, such as deet, on a treated net to form a barrier around a perimeter. 13 For more 
discussion of area repellents, see Chapter 23 by Strickman. 

At times, the term “spatial repellent” is used to describe the action of some pyrethroids. 14 ' 15 It should 
be noted that pyrethroids can produce excito-repellency with possible mortality as a result of the 
exposure. 16 A pyrethroid with sufficiently high vapor phase concentration, e.g., metofluthrin, 1517 can 
result in a spatial repellent (barrier) effect regardless of knockdown and mortality of insects. In this 
chapter, the discussion is mostly confined to the natural compounds that impact mosquito behavior by a 
means of masking attractive odors while minimizing concerns over the mode of action, such as the 
pyrethroids that exhibit excito-repellency and insecticidal properties. 


© 2006 by Taylor & Francis Group, LLC 


80 


Insect Repellents: Principles, Methods, and Uses 


Attraction-Inhibitors 

The term “inhibition” has been used to describe a net behavioral effect from a particular mechanism, 
such as “distension-induced inhibition .” 18 Simpson and Wright described the use of low-level 
continuous emission of a chemical, e.g. Rutgers 612 (2-ethyl-1,3-hexanediol) as a means to “inhibit 
the normal response” of mosquitoes to an increase in the carbon dioxide gradient . 19 Although the normal 
response to carbon dioxide can range from flight activation to oriented positive anemotaxis to the odor 
source, it is assumed that in this case, the authors expected the normal response to be that of attraction. 
The term “inhibitor” also denotes a compound that imparts a reduction in trap catches for traps baited 
with a pheromone . 20 ' 21 Davis linked a decrease in sensitivity of lactic acid receptor neurons to the 
inhibition of host-seeking behavior following a blood meal . 22 If we adopt and apply “inhibitor” in an 
analogous way to describe these allomones that inhibit the activity of kairomones, these compounds are 
then inhibitors of attractants (i.e. attraction-inhibitors analogous to pheromone inhibitors described by 
Roelofs and Comeau," and Kennedy)." Torretal. later expounded on the work of Davis and discussed 
the manner in which these “attractant-inhibitors” may affect the insects. We have shifted away from 
calling human-produced masking chemicals “spatial repellents” in recent years and adopted the term 
“attraction-inhibitors.” We believe this term to be a logical choice to describe the observed behavioral 
effect (inhibition) in bioassays . 24,25 


Spatial Repellency and Attraction-Inhibition Research 

In the mid 1960s, Skinner et al . 26 collected lipid fractions from human skin exudates and reported that 
some of these lipid fractions were “repellent” to mosquitoes. They hypothesized that the attraction of 
mosquitoes to humans was more complex than simply locating a host using kairomones only. It was 
speculated that the combination of human-produced kairomones and allomones resulted in the overall 
measured attractiveness of an individual to mosquitoes . 26,27 Further investigation of the lipid fractions 
implicated unsaturates as the repellent allomones . 28 Moreover, measured attraction increased when these 
lipids were removed from sweat. Skinner et al . 29 later identified the most repellent of these acids as 
a-linolenic (9,12,15-octadecatrienoic), 2-decenoic, 2-nonenoic, arachidonic (5,8,11,14-octadecatetrae- 
noic), and 10-undecenoic acids. Some saturated fatty acids, e.g., caproic (hexanoic), enanthic 
(heptanoic), and pelargonic (nonanoic) acids also exhibited high repellency. The carboxylic acids and 
their effect on host-seeking will be examined more in-depth in Section “Attraction-Inhibition by 
Carboxylic Acids”. 

Early History of Spatial Repellents Testing 

The notations of vapor phase, spatial, and area effects from repellents were reported by Christophers in 
1945, and this was especially noticed from the action of pyrethrins . 30 Christophers also noted a 
distinction between “contact" repellents and “vapour” repellents described by McCulloch and Water- 
house . 31 A concerted search for “spatial” repellents was undertaken by the USDA in 1948. 32 Although 
we will continue to use “spatial” in place of “vapour” repellents, they are both defined by the respective 
authors as repellents that work “at a distance .” 11,31 The USDA effort came about as an offshoot from the 
established program of topical repellent testing. In their first report of spatial repellents, some of the 110 
chemicals tested exhibited repellency, but none were deemed to be outstanding. Results based upon this 
USDA effort were first described in the literature by Gouck et al. 11 ; who reported on the spatial 
repellency of various esters using time of protection from bites as the means of quantifying the 
differences in repellency (see Section “Olfactometers for the Assessment of Spatial Repellents” for a 
description of their bioassay system). Of the esters that were tested, it was reported that the spatial 
repellency for the mandelates increased as a function of the carbon chain length from C 3 to Cs, with an 


© 2006 by Taylor & Francis Group, LLC 




Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


81 


optimum that occurred at C 5 . McGovern et al . 33 used a similar assay system to study other compound 
classes, and in particular noted that deet was an effective topical repellent, but performed only weakly as 
a spatial repellent. Maximum spatial repellency occurred in the C 5 to C 9 range for most of the compound 
classes tested. Later, other common topical repellents were examined and dimethyl phthalate (DMP) was 
reported as one of the best spatial repellents against Aedes aegypti 34 ', however, Khan and Maibach found 
deet to be better than DMP using their own biossay methods . 35 Other noteworthy substances that have 
been shown to produce spatial repellency of Aedes aegypti include essential oils like citronellal 
(3,7-dimethyl-6-octen-l-al) and geraniol (E-3,7-dimethyl-2,6-octadien-l-ol), pyrethrums and pyre- 
throids, and common topical repellents . 14 Recent efforts to find inhibitory chemicals are directed at 
natural compounds by examining differences among individual humans of variable attractiveness to 
biting arthropods. 

Spatial Repellency and Attraction-Inhibition of Catnip Oil 

It has been known for some time that volatiles produced from catnip, specifically the isomers of 
nepetalactone, repel phytophagous insects . 36 ' 37 Peterson and Coats examined catnip oil and nepetalac- 
tone isomers as alternatives to deet for protection from mosquitoes and found these to be more repellent 
than deet in their bioassay system . 38 ' 39 Recently, catnip was examined for its ability to inhibit the host¬ 
seeking of mosquitoes and was found to be a better attraction-inhibitor than deet, but the less effective 
repellent of the two based on mean complete protection time (CPT) on a treated cloth affixed to a card 
above the skin surface . 40 Further examination of catnip oil and its constituents to deter biting was 
conducted by Chauhan et al . 41 The results of their in vivo and in vitro studies were similar in that the 
biting deterrency of each of the two nepetalactone isomers (Z,E- and E,Z-) and of the racemic mixture 
were all significant compared to the control, but not different from each other. Tested in vitro, these 
compounds did not deter biting as well as deet or the repellent, (lS,2S , )-2-methylpiperidinyl-3- 
cyclohexene-1 -carboxamide (SS220). Further discussion of natural plant and botanical insect repellents 
are the topics of Chapter 14 by Moore and Hill, and Chapter 15 by Gerberg and Novak, respectively. 

Spatial Repellency and Attraction-Inhibition of Deet 

Deet has produced mixed results as a spatial repellent as was mentioned briefly in Section “Early History 
of Spatial Repellents Testing” and Section “Spatial Repellency and Attraction-Inhibition of Catnip Oil”. 
In some cases, it is weak or less effective than other compounds , 11 ' 33 34 40 42 and in others it is more 
effective . 14 ' 35 ’ 43 One possible explanation is that the concentration of deet needs to reach a specific 
threshold in the vapor phase so that the concentration is sufficiently high enough to affect the mosquito 
chemosensilla. Otherwise, vapor phase concentrations below this level require landing on a topically 
treated surface to result in contact with deet at sufficient concentration to act as a biting deterrent and 
therefore be repellent. Dogan et al . 25 concluded that deet inhibited the action (attraction) of L-lactic acid, 
but did not act as a repellent. Dogan and Rossignol noted that just after topical application of deet , 24 test 
subjects were still attractive to Aedes aegypti. The results of Bernier et al 40 showed that deet inhibits the 
attraction of mosquitoes, but when compared directly at equivalent dosages, it did not function as an 
attraction-inhibitor as effectively as catnip oil. It merits mentioning that the individual volunteer whose 
odors were used in the Bernier et al. study was relatively less attractive to Aedes aegypti than most 
individuals and this may have produced atypical results. Normally, the mixing of deet into the air stream 
of a port with human odors does produce a small decrease in the percentage of mosquitoes collected in 
the olfactometer trap . 44 

Dogan et al . 25 reported deet to be attractive in the absence of L-lactic acid; this has been reported 
previously for low doses of deet and Rutgers 612 45 We have also observed this in bioassays with our 
olfactometer . 46 In the absence of attractive odors, the clean airstream in our system produces no response 
(no flight activation nor positive anemotaxis) by the mosquitoes. However, with the release of a 


© 2006 by Taylor & Francis Group, LLC 


82 


Insect Repellents: Principles, Methods, and Uses 


chemical, upwind anemotaxis and subsequent trapping of a few mosquitoes is observed at times, even 
when a compound does not attract mosquitoes when tested in competition against a potent attractant 
(e.g., host odors or chemicals based on host odors ). 44 What appears to be important, at least in the case of 
deet, is that wind movement contributes significantly to the ability of a compound to repel (or perhaps 
more appropriately, inhibit host finding) in the vapor phase in both a controlled setting and in the 
field 43 ' 44 ' 47 

Attraction-Inhibition by Linalool and Related Compounds 

Alcohols are widely known to repel mosquitoes. For example, citronellol (3,7-dimethyl-6-octen-l-ol), 
and it’s related aldehyde analog, citronellal mentioned in Section “Early History of Spatial Repellents 
Testing”, exhibit spatial repellency of Aedes aegypti in laboratory bioassays . 14 In fact, essential oils, e.g., 
citronella Cymbopogon nardus, were the most commonly used repellents prior to the 1940s. 37 
Interestingly, citronella oil contains primarily geraniol; however, citronellol and citronellal were reported 
as the active ingredients leading to repellency . 48 ' 49 Linalool (3,7-dimethyl-l,6-octadien-3-ol) is a water- 
insoluble alcohol that is a colorless liquid and is used commonly by the perfume and cosmetics industry 
because of its appealing flowery odor. It can be found naturally in such sources as apricots, carrots, 
lavender, cardamom and marjoram. Human inhalation of this compound is known to produce sedation, 
and it has been shown to suppress the voltage-gated currents in newt olfactory receptor cells . 50 Birkett et 
al . 51 reported that linalool produced significant electroantennogram (EAG) responses in four species of 
biting flies, and reduced the upwind (positive) anemotaxis in laboratory wind tunnel studies. 

Linalool has two optically active isomers; researchers have found the (S)-( + )-enantiomer to be the 
better attraction-inhibitor . 52 Using a dual-port triple-cage olfactometer , 46 Kline et al. examined the 
impact of linalool, dehydrolinalool (3,7-dimethyl-6-octen-l-yl-3-ol), and deet on the host-seeking ability 
of laboratory-reared Aedes aegypti. 53 Compared to dehydrolinalool and deet in competitive bioassays, 
linalool was the most potent inhibitor (competitive bioassays are defined in Section “Considerations in 
the Experimental Design”). An important finding of this work was that the release of linalool resulted in 
two observable effects on mosquito behavior. The first effect was that fewer mosquitoes in the cage were 
activated to flight from concomitant release of attractant plus linalool in the airstreams of separate ports 
of the dual-port olfactometer. This indicated that vapor phase linalool acted as an attraction-inhibitor by 
preventing some of the mosquitoes from detecting the normally attractive odors. The second observable 
effect was that of the mosquitoes that were activated to flight, fewer than normal numbers of these were 
able to locate the odor source. This indicated that even though some mosquitoes could detect the presence 
of attractive odors, they were not as capable of orienting towards and, thus, locating the odor source. 

Human-Produced Compounds That Affect Host-Seeking 

The skin surface of humans differs greatly from that of other animals. Except for a few specific localized 
areas, human skin normally ranges from pH 4.2-6.0 due to the abundance of fatty acids that are 
present . 54 " 55 In addition to carboxylic acids, skin also has high levels of triglycerides and squalene; 
however, it is the acids that contribute largely to the types of microbes that can exist on skin . 56 " 57 Humans 
are the only animal to exhibit acne vulgaris, and within the comedo of acne, there are high levels of fatty 
acids . 58 The distribution of saturated fatty acid molecular sizes are clustered in the C 12 -C 20 and C 21 -C 30 
ranges, of which Ci 6 and C ]8 in the former and C 24 in the latter are present in the highest relative 
abundance . 59 The most abundant unsaturated fatty acids are palmitoleic (9-hexadecenoic), oleic 
(9-octadecenoic) and linoleic (9,12-octadecadienoic) acids. 

While some studies have focused on endogenous lipid production, others have focused more on the 
end products, or volatiles that are released by metabolic activity via respiration through the skin, or from 
degradation of skin surface compounds by microbial action. Sastry et al . 60 assembled a comprehensive 
treatise of human-produced compounds, covering how these compounds can be used in the diagnosis of 


© 2006 by Taylor & Francis Group, LLC 


Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


83 


diseases and in the interpretation of human metabolism. In their review of the subject, they highlighted 
studies that identified chemically underivatized compounds, such as that of Ellin et al. who reported the 
identification of over 130 compounds in a study of total human body emanations. 61 Among the high 
levels of acids, there were also significant volatile constituents identified that consisted of alcohols, 
ketones, aldehydes and other chemical compound classes. For example, acetone and isoprene (2-methyl- 
1,3-butadiene) were two of the most abundant components, emitted at rates of 240^-70 and 251-425 pg/h, 
respectively, for the three subjects that were examined by Ellin et al. 61 In a later study, Naitoh et al. 
determined the release rate of acetone from human skin and reported a range as 80-800 pg/cm 2 min. 62 In 
our own studies, we knew of one individual (volunteer A in Table 4.1) who consumed alcohol regularly. 
This subject was consistently the most attractive to Aedes aegypti of the six subjects who participated in 
this study. Of the most volatile emanations quantified, the most attractive individual (A) produced the 
highest level of acetone, ethanol, and methanol. Shirai et al. 63 reported that landings of the Asian tiger 
mosquito, Aedes albopictus increased after consumption of a beverage containing ethanol. They 
measured both skin temperature and the ethanol in the perspiration of human subjects, but they did 
not find a relationship between either of these two variables and the landing rates. 

Similar to Sastry et al. 60 we also prefer the analysis of volatiles without chemical derivatization for the 
identification of human skin (and other host) emanations that may affect mosquito host-seeking. 64 
Mosquitoes detect volatile host-finding cues in the gas phase, so we believe that minimization of 
complexity in the sampling process will tend to cause the least change or bias toward the compound 
classes and proportions of each chemical detected. The relative abundances of many of the volatile 
compounds have a significant impact on the overall attraction process. Additionally, it is important to 
avoid comparing too closely human and mosquito olfaction. The odor of human perspiration that we 
smell is due inpart to saturated and unsaturated Cg-Cn acids and one of the most abundant odiferous 
compounds is (E)-3-methyl-2-hexenoic acid. 60 ' 65 ' 66 The sensitivity of humans to the odor of these 
compounds does not necessarily imply that these same compounds have a role in mosquito host finding. 


TABLE 4.1 


Comparison of Volatile Compounds Emanated from Six Different Humans 





Human Subject 




Compound 

(Class) 

A 

B 

C 

D 

E 

F 

(Aldehydes) 

Acetaldehyde 

160 

83 

74 

52 

190 

172 

2-Methyl-2- 

2.2 

1.4 

6.5 

1.5 

8.3 

7.8 

propenal 

2-Methylbutanal 

2.5 

1.3 

0.92 

0.87 

4.6 

4.7 

Hexanal 

5.4 

6.2 

8.1 

6.4 

29 

38 

(Ketones) 

Acetone 

900 

50 

24 

45 

168 

200 

2-Butanone 

1.3 

0.30 

0.39 

0.33 

1.0 

0.82 

2,3-Butanedione 

2.2 

1.4 

6.5 

1.5 

8.3 

7.8 

(Alcohols) 

Methanol 

638 

6.9 

4.1 

8.1 

13 

13 

Ethanol 

638 

219 

4.1 

117 

18 

45 

(Sulfides) 

Carbon disulfide 

0.38 

0.12 

0.13 

0.33 

1.7 

0.55 


The headspace of forearm emanations was collected in a Tedlar bag and analyses conducted by microscale purge and trap 
GC/MS. Values are in parts per billion by volume (ppbv). 


Source: From M. M. Booth, Unpublished results, 1997. With permission. 


2006 by Taylor & Francis Group, LLC 




84 


Insect Repellents: Principles, Methods, and Uses 


Bernier et al. 67 used gas chromatography/mass spectrometry (GC/MS) of compounds adsorbed and 
then thermally desorbed from glass beads to identify 277 compounds present on the skin of humans. 
They used columns with different stationary phase polarities to perform the chemical separation of 
samples collected from four subjects (males ranging from 26 to 61 years in age). The composition of 
emanations was qualitatively similar for all subjects, but quantitative differences were readily observed. 
This study provided the groundwork to explore chemical differences between individuals who 
represented the extremes of low and high attractiveness to biting mosquitoes. The same study 
also examined day-to-day chemical changes correlated to changes in laboratory measured mosquito 
attraction for a single individual. 68 In the comparison of two different subjects, the individual who was 
more attractive to Aedes aegypti had, on average, higher levels of lactic acid, butanone, 2-pentanone, 
3-pentanone, and 6-methyl-5-hepten-2-one. The less attractive host had a higher level of methylpentanol, 
1,3-butanediamine, capric acid (decanoic acid), lauric acid (dodecanoic acid), heptanal, and pelargo- 
naldehyde (nonanal). From studies of a single individual, nonanal, 6-methyl-5-hepten-2-one, and 
benzaldehyde were less abundant in the emanations on the day that the residuum was more attractive 
to mosquitoes. Those individuals who were less attractive to Aedes aegypti tended to have the highest 
concentrations of aldehydes, particularly nonanal, on their skin. 68 ' 69 Thus, aldehydes appear to have an 
important role in the balance of attraction and inhibition. 

Human fingerprint residues have been examined to identify gender-specific and age differences in the 
lipids. 70 Hexadecenoic, palmitic (hexadecanoic), and octadecenoic acids were among the most abundant 
acids observed, in agreement with Ansari et al. 59 Although these three acids occur at higher relative 
abundances in males compared to females, the differences were not statistically significant. Curran 
et al. 71-72 examined male and female odors over time and after exercise. They described a classification of 
detected compounds based upon the origin of the odors. “Primary odors” were comprised of emanations 
that were present regardless of sampling date or time. Compounds that originated from dietary or 
environmental factors were considered secondary odors. Tertiary odors were those attributable to 
exogenous factors that resulted in adherence of a chemical to the outer layer of the skin. Using this 
terminology, the base attraction of mosquitoes to human hosts would be associated with the primary odor 
components, with some differences possibly found in the secondary odors and less likely in the tertiary 
odors. Finally, before focusing on the specific compound classes in human emanations, it is interesting to 
note the similarities of constituents for skin compounds compared to those found in the oral cavity, urine 
and alveolar breath. 

Oral odors are comprised primarily of sulfides, ethanol, diacetyl (2,3-butanedione), acetone, 
acetaldehyde, and methyl mercaptan (methanethiol). 60 Acetaldehyde and other aldehydes are also 
detectable in blood and breath. 73-74 Many of the short-chain ketones, acids, and hydroxy acids, such as 
L-lactic acid, are also present in human urine. Breath has been reported to contain hundreds of detectable 
compounds, 75 ’ 76 and many of these constituents overlap with those present in blood, 60-77 urine, 60 and on 
the skin. 67-72 It is fairly obvious that exhaled breath contains large quantities of carbon dioxide and this is 
one of the most universally known behavioral activators and trap attractants for mosquitoes. 78-82 
However, breath also contains compounds that inhibit the host-seeking response, as was shown for 
Anopheles gambiae 83 Therefore, in addition to known Aedes aegypti attractants (e.g., acetone, dimethyl 
disulfide (DMDS), and 2-pentanone) in human breath, there are also attraction-inhibitors, e.g., 
nonanal. 75-76 From attraction studies, it was evident that certain combinations of chemicals and 
classes of chemicals when combined with L-lactic acid resulted in blends with much lower than expected 
attraction of mosquitoes in laboratory bioassays. ’ ’ For example, in Bernier et al. some branched 
ketones and aldehydes that were combined with L-lactic acid resulted in attraction responses that were 
less than that of L-lactic acid alone (26% in this study ). Some of these specific compounds and functional 
groups are discussed in greater detail in the next few subsections. While generalizations can be made 
about specific compounds and their ability to attract and inhibit, it is crucial to keep in mind that the 
reported behavioral effect is heavily dose-dependent. Specifically, some compounds that attract at low 
vapor-phase concentrations may inhibit, arrest, or repel insects at higher concentrations [viz. the response 
of deet described in Section “Spatial Repellency and Attraction-Inhibition of Deet”]. 


© 2006 by Taylor & Francis Group, LLC 


Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


85 


Attraction-Inhibition by Carboxylic Acids 

Saturated and unsaturated fatty acids are abundant in skin emanations. Other substituted acids such as 
L-lactic acid are also present at relatively high levels, and dicarboxylic acids can be detected as some of 
the constituents deposited on handled glass beads . 67 L-lactic acid is a hydroxy acid that is expected to be 
present at substantial levels because it is formed in the body from metabolism of proteins and 
carbohydrates under anaerobic conditions. Ellin et al. also detected another important metabolic 
product, pyruvic acid, which is an oxo acid that plays a vital role in human metabolism under aerobic 
conditions . 61 

The initial discovery that fatty acids resulted in “repellency” (inhibition of host-seeking in 
bioassays) was reported by Skinner et al. as noted in Section “Spatial Repellency and Attraction- 
Inhibition Research ” 26 Examination of the volatile acids used in blends developed to attract Aedes 
aegypti led to discoveries about the compounds that inhibit host-seeking, specifically that addition of 
some saturated acids to blends decreased the attraction. Bosch et al. used a Y-tube olfactometer to 
demonstrate that combinations of L-lactic acid and either butanoic (C 4 ) or any of the C 9 -Q 2 acids 
resulted in a composition that did not produce a significant increase in the attraction of female Aedes 
aegypti compared to the attraction to L-lactic acid alone . 86 For binary blends of L-lactic acid and either 
propanoic (C 3 ) or pentanoic (C 5 ) acids, they observed that addition of either undecanoic (Cn) or 
myristic (tetradecanoic) (C 14 ) acids to this blend resulted in a significant decrease in attraction. 
Smallegange et al. reported that a blend of 12 carboxylic acids was repellent against Anopheles 
gambiae when tested alone or with L-lactic acid . 87 

Constantini et al. reported that the electrophysiologically active acids that produce the odor in human 
perspiration, such as Z- and E-3-methyl-2-hexenoic acid, and 7-octenoic acid, repelled or masked the 
presence of attractants, and that these may be involved in the avoidance of nonpreferred individuals for 
blood meals . 88 These findings and those of Bosch et al . 86 provide compelling support to the view that as 
the concentration of constituents in the human odor profile is perturbed greatly, it can result in host- 
avoidance behavior by mosquitoes. 

Reifenrath indicated that acids in the C& to C 8 range coupled with C 8 -C 12 acids were repellent to 
arthropods, and that binary combinations of octanoic (C 8 ) and nonanoic (C 9 ) acids, or the tertiary 
combination of C 8 -C 10 acids effectively prevented host location . 89 Reifenrath examined repellency of 
Aedes aegypti by treating gauze or polyester film with each acid applied at 0.3 mg/cm 2 . These 
experiments indicated that 2-pentenoic, 2-octenoic, 3-methyl-2-octenoic, nonanoic, decanoic, and 
undecanoic acids were the most effective. Topical tests on human skin showed that the most 
repellent compounds were 4-methyloctanoic, 3-methyl-2-octenoic and nonanoic acids, implicating the 
most repellent compounds as those that contain 9 carbons and to some extent 8 and 10 carbons 
[viz. nonanal discussed throughout this chapter, but also the C 8 and C 10 carbon compounds such as 
linalool, citronellol, citronellal, dehydrolinalool in Section “Spatial Repellency and Attraction- 
Inhibition of Deet” and Section “Attraction-Inhibition by Linalool and Related Compounds”, geraniol 
in Section “Early History of Spatial Repellents Testing”, and Z-4-decenal and octanal in Section 
“Attraction-Inhibition by Aldehydes”]. 

Attraction-Inhibition by Aldehydes 

Aldehydes have received attention recently because they have been identified as the repellent compounds 
in the emanations of the crested auklet (Aethia cristatella). 90 ~ 93 Three of the four reported repellents are 
aldehydes, hexanal, octanal, and Z-4-decenal, and one is an acid, hexanoic acid. Additional discussion of 
these compounds and chemical defenses of birds and other vertebrates is found in Chapter 3 by Weldon 
and Carroll. Although nonanal was not identified in emanations of the crested auklet, it has been reported 
as a major constituent of emanations from the whiskered auklet ( Aethia pygmaea). 91 Nonanal appears to 
be detected not only by mosquitoes, but other blood-feeding arthropods as well. Steullett and Guerin 
demonstrated that numerous aldehydes, including hexanal, heptanal, nonanal, benzaldehyde, and 
methyl-substituted benzaldehydes stimulated tarsal chemoreceptors of the tick Amblyomma variegatum. 


© 2006 by Taylor & Francis Group, LLC 


86 


Insect Repellents: Principles, Methods, and Uses 


another arthropod that relies at least in part on chemical cues for host location . 94 Guerenstein and Guerin 
identified nonanal as the compound that elicited an electrophysiological response from a receptor on the 
basiconic sensillum of triatomine bugs ( Triatoma infestans ). 95 In that study, nonanal was also identified 
chemically by GC/MS in the extracts of sheep wool and chicken feathers. The unsaturated and 
diunsaturated aldehydes tested in their study did not produce an electrophysiological response, nor 
did other Cg compounds, including nonanoic acid, 2-nonanone, and nonanol. Heptanal and octanal also 
produced linear responses in the sensillum cells, but other saturated aldehydes (C 6 , C 10 -C 12 ) did not. 
Interestingly, researchers have previously observed a linear correlation of attraction and repulsion to the 
concentration of aliphatic aldehydes in blowflies . 37 ' 96 

Aldehydes are commonly reported in residue from human skin; these are predominantly in the C 6 -C| 0 
range . 72 Haze et al. documented that the concentration of 2-nonenal, an unsaturated analog of nonanal, is 
related to the age of an individual with higher levels observed in males over 40-year-old and that all 
subjects produced detectable quantities of C 6 -C 10 saturated aldehydes in this study . 97 In contrast, Curran 
et al. was able to detect 2-nonenal in females and in individuals less than 25-year-old. 71 ' 72 Curran et al. 
reported that the C 8 -C 10 aldehydes were detectable in 88% of their subjects , 71 and Zhang et al. also 
reported these C 8 -C ]0 aliphatic aldehydes . 98 A better understanding about the role of C 8 -C 10 aldehydes 
in the mosquito host-finding process may benefit from experiments comparing the relative attractiveness 
of subjects who have high or low concentrations of these compounds on their skin. 

Bernier et al. used microscale purge and trap GC/MS to identify aldehydes from butanal to 
undecanal, with nonanal as the most abundant in this series . 64 The cryo-focused GC/MS analysis of 
glass beads allowed the detection of propanal (C 3 ) to nonanal (C 9 ), including branched and 
unsaturated analogs of these compounds. The more volatile aldehydes are partly responsible for off 
odors in spoiled meat , 99 while the less volatile, such as octanal, nonanal, and benzaldehyde have a 
more pleasant floral aroma. Endogenous aldehydes that are oxidized from their respective acids are 
hexanal from linoleic and arachidonic acids; heptanal from palmitoleic acid; and nonanal from oleic 
acid . 74 ' 100 As noted earlier in this chapter, these acids are the some of the most abundant in human 
emanations . 59,67,70 By analogy, this may partly explain the abundance of these specific aldehydes in 
human emanations. 

Attraction-Inhibition by Ketones 

Acetone is the most abundant ketone in human odors (see Table 4.1 ). 61,62 One mechanism for 
endogenous production of this compound is from fat metabolism . 62 In addition to acetone, numerous 
2- and 3-substituted ketones, as well as cyclohexanone, have been reported in human odors . 61 
Unsaturated ketones have also been found in the residue of more than 50% of human subjects . 72 
Birkett et al . 51 reported that when the unsaturated branched ketone, 6-methyl-5-hepten-2-one, was 
applied to cattle, it reduced the attraction to biting flies. 

Saturated ketones, particularly in the C7-C12 range have been found to inhibit mosquitoes . 101 The 
combination of L-lactic acid with either acetone or butanone, the smallest and most volatile of the 
saturated ketones, produced synergistic attractant blends for Aedes aegypti . 101,102 However, as larger 
saturated ketones within the series, like pentanone (C 5 ) and hexanone (C 6 ), are blended with L-lactic acid, 
the attraction drops from synergistic to additive, and then results in inhibition of attraction for blends with 
heptanone (C 7 ) through dodecanone (C 12 ). When chain lengths exceed C ]2 in the ketones (C ]0 in acids 
and aldehydes) it is expected that the volatility decreases below a threshold level such that the vapor 
phase concentration is so low that the impact on host-seeking disappears. This effect was also evident 
when researchers examined the repellency of alcohols larger than decanol . 37 

Attraction-Inhibition by Alcohols 

Bernier et al . 67 identified unsaturated and saturated alcohols from butanol to heptadecanol were in human 
skin. Ellin et al. also observed a number of these alcohols and ethylene glycol . 61 Glycerol also was 


© 2006 by Taylor & Francis Group, LLC 


Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


87 


reported in both studies; it is a major breakdown product of bacterial action on triglycerides. 56 Phenol 
was produced by all human subjects in the study of Curran et al. 72 In addition to amides like deet, 
aliphatic alcohols have been popular historically as insect repellents, e.g., the series of decanol (Ci 0 ) 
through tetradecanol (Ci 4 ), 103 and Rutgers 6 1 2. 104,105 Dogan and Rossignol examined various fragrances 
and compositions that contained alcohols such as geraniol and dimethyl cyclormol (hexahydrodimethyl 
methanoinden-5-ol) and found these to either inhibit or repel mosquitoes in a modified Feinsod- 
Spielman olfactometer. 24 

In contrast to the well known attractant l-octen-3-ol, 106,107 several related, more volatile unsaturated 
alcohols, including linalool will inhibit attraction by Aedes aegypti in laboratory bioassays. 53,108 Yet, 
other unsaturated alcohols, such as geraniol, 24 or diols that are similar in structure, such as 7-octen-l,2- 
diol, have little or no effect on the host-seeking of Aedes aegypti. 108 The examination of compounds from 
cattle to identify compounds that affect host location by five species of biting flies revealed that 1-octen- 
3-ol and 3-octanol were attractants in wind tunnel studies. In contrast, these compounds reduced the 
number of biting flies on cattle in the field. 51 This may be a case where the normal host odor profile is 
perturbed so greatly by the added volatiles that host avoidance by the insects is the net result. 

Attraction-Inhibition by Compounds of Other Classes 

Researchers have documented ammonia and a series of amines from methylamine to butylamine in 
human emanations. 61,109 Ammonia is formed through amino acid catabolism, and along with urea and 
uric acid are the three main nitrogen-containing compounds excreted by animals. 110 Ammonia has been 
demonstrated to attract Aedes aegypti and Anopheles gambiae at low concentrations, 111,112 and to deter 
feeding at higher doses. 87,113 In addition to these alkaline substances, Bernier et al. also reported a 
substantial number of hydrocarbons and heterocyclic compounds present in human emanations. 67 Some 
of these are currently being tested in our laboratory to determine if they play a role in the host-seeking 
behavior of mosquitoes. Bernier et al. identified some sulfides and some 1-chloroalkanes in human skin 
emanations. 67 Sulfides and chlorides have not been observed to inhibit the host-seeking of Aedes 
aegypti 101 ; however, larger sulfides, chlorides and other alkyl halides have not yet been tested as 
attraction-inhibitors. 

If we attempt to make a general statement regarding compounds capable of attraction-inhibition, then 
we could base this upon the presence of oxygen in the molecule, as Bunker and Hirshfelder noted for 
“good” repellents in 1925. 114 Roadhouse later noted that many effective repellents contain nitrogen. 115 
However, this should be kept in perspective because many compounds contain oxygen, nitrogen, or both 
and do not show effective repellency or inhibition of mosquito host-seeking. 115,116 


Identification of Host-Produced Allelochemicals 

Numerous techniques exist to sample, collect, concentrate, chemically separate, and identify compounds 
in host emanations. There are benefits and drawbacks to each choice. One needs to consider all of these 
factors carefully when selecting the approaches to solve a complex problem, such as the identification of 
chemicals that affect mosquito host-seeking behavior. It is important to realize that a single method in 
any of these processes is likely to prove inadequate for the resolution of a complex situation involving 
potentially numerous compounds that can span a wide range of differing compound polarities and 
volatilities. For example, multiple preconcentration techniques may be needed to provide comp¬ 
lementary information, and multiple chromatography columns with stationary phases of different 
polarities may need to be used to resolve all of the compounds. 64,67 By combining information from 
different types of analyses, the total chemical profile will be more complete. Some of the more recent 
techniques applied to the analysis of human emanations have either involved solvent extraction. 


2006 by Taylor & Francis Group, LLC 



88 


Insect Repellents: Principles , Methods, and Uses 


deposition onto glass beads, or the use of solid phase microextraction (SPME) fibers as noted in Section 
“Analysis of Human Emanations”. 

Analysis of Human Emanations 

The analytical method of choice for almost all comprehensive chemical analyses of volatile human body 
emanations has involved chromatographic separation followed by mass spectrometric detection, e.g., 
GC/MS, whether the emphasis is on skin emanations, breath, urine, blood, oral cavity, or the total 
composite of emanations from an entire person. 60 ’ 61 ' 64 ’ 67 ’ 68 ’ 70-72,75,76,98,t09,i 17,1 is jy[ ass spectrometry 
allows for the identification of compounds based on the fragmentation pattern of compounds. These 
patterns consist of differing intensities of ions (technically, as a ratio of mass to charge, m/z) that result 
from bombardment of sample molecules by electrons. There are various types of mass analyzers for mass 
spectrometers, but the most common for these studies are either magnetic/electric sector or quadrupole 
instruments because they provide mass spectra that is most similar, and therefore the most easily 
matched, to mass spectra in existing computerized mass spectral libraries. 

In many of these analyses, hundreds of compounds are present. Therefore, separation must be 
effected prior to mass spectral analysis. This is accomplished by column chromatography. Over the 
last few decades, the columns employed for this purpose have improved greatly. They are more stable 
due to better phase bonding, allow greater sample capacity, and are capable of better resolution. 
Despite all of these improvements, exposure to air and/or extreme hot or cold temperatures still easily 
degrade the GC column stationary phase. In general, the more polar that the column phase is, the more 
constrained that it will be with respect to temperature limits than a column that has a relatively 
nonpolar stationary phase. 

Soxhlet extraction, commonly used for fat and oil extraction, followed by GC/MS was used to 
concentrate and identify volatiles from foot stockings. 119 Bernier et al. used glass beads to collect 
emanations for subsequent thermal desorption into a GC injection port. 64 ' 67,68 In doing so, the problems 
from the high water content of perspiration was avoided. This remedy is significant because loading 
water onto gas chromatography columns is detrimental to the stationary phase. Asano et al. used glass 
beads followed by solvent extraction of compounds from the beads to study fingerprint residues. 7 
Headspace GC/MS was used to analyze age-specific male individual odor differences, 97 and SPME has 
been used to collect and concentrate skin volatiles for subsequent identification and quantitation by GC/ 
MS. In the work of Curran, supercritical fluid extraction (SFE) was used as a pretreatment to 

reduce or eliminate some of the background compounds in the gauze, which was necessary for 
quantitation of human emanations because a number of human emanations also are measurable in the 
background contaminants from the gauze. This innovative pretreatment reduced exogenous compounds 
and allowed them to achieve accurate quantitative results. 


Merging Chemistry and Sensory Physiology 

One of the earliest reports of detection of electrical impulses along the nerves was that of Adrian, who in 
1930 recorded the discharge of the caudal nerve in the caterpillar. 1 ' 0 Electrophysiological studies of 
these impulses based upon selection of innervated nerve has contributed significantly toward an 
understanding of which compounds and which sensory organs may factor into the process of host 
attraction or other behavioral responses. Electroantennograms provide an ideal screening tool for 
compounds that insects detect, although it does not reveal whether this detection may lead to attraction, 
avoidance, repellency, or other behaviors. Single-cell recording can determine precisely which receptor 
organ a compound stimulates. In the early days of these techniques, Roelofs used GC to separate 
compounds and coupled the resulting sample stream with EAG to identify pheromones and compounds 

'sa 101 1 77 

that are synergists and inhibitors for pheromones.' ’ 


© 2006 by Taylor & Francis Group, LLC 



Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


89 


Combination of these techniques with gas chromatographic separation is a powerful approach to 
analyze complex samples for the compound (peaks) that produce bioactivity. By either routing the 
sample via column splitting to both instruments (GC-EAD and GC/MS), or simply injecting the same 
sample on separate instruments with the GCs configured similarly, it is possible to identify and thus focus 
on a smaller set of compounds that are bioactive in a sample that may contain hundreds of compounds. 
Recent applications of this technique can be found in the report of Cork et al. 123 and related studies 
involving mosquitoes, such as Anopheles gambiae. 117 A24 ~' 26 Working with Anopheles gambiae 
antennae, Cork and Park examined extracted human skin compounds and identified the most abundant 
acids as acetic, heptanoic, and hexadecanoic acids, whereas the EAG responses were greatest for formic, 
pentanoic, butanoic, propanoic, acetic, and hexanoic acids, all of which were more intense than the 
response to the l-octen-3-ol standard. 117 Constantini et al. 88 examined EAG responses of common 
human-produced odiferous compounds in sweat and evaluated their impact on host-seeking using a wind 
tunnel for bioassay as reported in Section “Attraction-Inhibition by Carboxylic Acids”. Other successful 
recent electrophysiological studies with additional arthropods have been reported for tsetse flies, 106 ' 127 
ticks, 94 and the New World screwworm. 128 

Current State and Future Directions of Host Odor Research 

Section “Attraction-Inhibition by Carboxylic Acids” described a recent example of the application of 
allomonal odors in which Reifenrath added carboxylic acids to host emanations to make the normally 
attractive host appear to have a different chemical profile. 89 The result was that the host was much less 
attractive to biting insects. At present, host-odor research continues with increased emphasis on 
understanding how kairomones and allomones function together to mediate the overall behaviors in 
the host-seeking process of arthropods. Some of the studies involve human hosts for anthrophophilic 
species that transmit malaria, such as Anopheles gambiae and Anopheles albimanus , or for those that 
transmit dengue and yellow fever, such as Aedes aegypti. Other studies center on birds, the preferred 
hosts of ornithophilic species such as Culex tarsalis, Culex pipiens quinquefasciatus and Culex 
nigripaplus, which are vectors of West Nile Virus (WNV) in North America. Studies involving 
animals as sources of chemicals that may attract arthropods, repel them, or inhibit the attractive 
emanations of a host is the subject of Chapter 3 by Weldon and Carroll. 


Laboratory Bioassays of Spatial Repellents and Attraction-Inhibitors 

The information derived from a particular study depends heavily upon the bioassay because the 
construction design of the device and the procedure used determine the behaviors that are assessed. 
The subject of this section is the common laboratory bioassay devices that have been used to produce 
many of the results described in this chapter. Additional coverage of olfactometer design and usage can 
be found in Chapter 9, written by Butler. 

Olfactometers for the Assessment of Spatial Repellents 

One can trace the design of dual-port olfactometers back to the 1930s. 129-131 Early USDA spatial 
repellency studies employed a similar style single-cage olfactometer modified to hold mesh netting in the 
trap ports. ’ Researchers conducted tests by passing air over a human arm and through a trap into the 
cage where 100 mosquitoes were located. The mesh cotton netting within the traps was either treated or 
untreated (as the control) with candidate spatial repellents. The test period was 5 min and netting was 
tested every other day until two successive trials resulted in > 10% of the test mosquitoes trapped in the 


© 2006 by Taylor & Francis Group, LLC 




90 


Insect Repellents: Principles, Methods, and Uses 


port with human odors. Thus, effectiveness of compounds was evaluated based on days of duration 
of repellency. 

Skinner and colleagues also used a dual-port olfactometer, operated in noncompetitive and 
competitive modes (see Section “Considerations in the Experimental Design” for a description of 
these modes) to compare two treatments consisting of human lipid fractions. 26-29 Researchers also 
quantified the repellency by the location of mosquitoes in the test and control ports after allowing the 
insects to fly upwind and select a port. In the noncompetitive mode, they compared the ratio of 
mosquitoes captured in the control port to the number in the sample port and the greater the ratio, the 
higher “repellency” according to this index. Because this experiment did not allow contact between 
mosquito and attractant, we believe that operation of the olfactometer in this way measured the 
attraction-inhibition of specific compounds. Dogan and Rossignol modified a Feinsod-Spielman 
olfactometer by constructing an additional chamber to allow measurement of “repellent” response 
based on insects moving away from the treatment. 24 Recently, Grieco et al. 134 designed a modular 
bioassay device which can be assembled to provide a system to screen contact irritancy of candidate 
chemicals, and reconfigured in a manner to allow assessment of spatial repellency. The movement of 
chemical inside each of these olfactometers is accomplished by convection and diffusion, without 
supplementation of a stream of air. 


Olfactometers for the Assessment of Attraction-Inhibitors 

Barrows first used the Y-tube olfactometer in studies of flies. 135 Geier et al. and Bosch et al. have 
used recent models to test mosquito responses. 86 ' 136 The triple-cage dual port olfactometer constructed 
by Posey et al. 46 and used in our laboratory is based on older designs described in Section 
“Olfactometers for the Assessment of Spatial Repellents”. ’ Because all of these olfactometer 
designs employ two ports, they can be used to measure attraction response to either a single treatment 
versus a control, or to two individual treatments in competition. Reifenrath used a Feinsod-Spielman 
olfactometer to measure the repellent effect imparted by carboxylic acids on human odors. 89 The 
design of this olfactometer allowed odors to pass through a linear arrangement (similar to Grieco 
et al. 134 ) of chambers by (in this case) a fan that drew the odors upward into the top chamber. Prior to 
conducting a test with human odors, mosquitoes were released in the top chamber, and allowed to 
distribute between the two chambers. After human odors were introduced through the bottom of the 
olfactometer, the mosquitoes that flew from the upper chamber down to the lower chamber were 
counted as responding to an attractive stimulus. Those remaining in the upper chamber were 
considered “repelled.” Again, this may not be truly indicative of repellency—it can be reasonably 
argued that mosquitoes that remain in the top part could be inhibited from detection of potentially 
attractive odors, or simply nonresponding. Provided that a standard is assessed with this design, then a 
reduction in attraction can be attributed to either the effect of a spatial repellent or attraction-inhibitor. 

As noted above, the standard design of the Y-tube, or dual-port olfactometer (without modification 
inside the traps) is perhaps not the best bioassay system to measure spatial repellency because one 
cannot discern whether mosquitoes left in the original position were nonresponding or truly 
“repelled.” Additionally, it remains unclear how to characterize mosquito behavior response to a 
treatment when they respond by positive anemotaxis into the clean air (control) port. For occasions 
that we observe this phenomenon, we always follow the test by examining the response to the 
individual control apparatus with no treatment in the opposite port to test for contamination of either 
the port or apparatus. 

Considerations in the Experimental Design 

In a dual-port olfactometer, there are two common modes by which the device can be operated and this is 
based on the number of treatments. A noncompetitive assay is arranged so that there is a treatment in one 


© 2006 by Taylor & Francis Group, LLC 


Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


91 


port compared to a second port containing the blank control (all apparatus used to hold the treatment, but 
with the treatment absent). 53 The advantage of this mode of operation is that it allows a means to compare 
attraction to treatments based upon a measurement of "inherent” or “absolute” attraction level, without 
possible interference or complications in mosquito behavior that may arise from interaction with odor 
released from a treatment in the second port. This approach is commonly used in our laboratory to screen 
for attractants and inhibitors. 

In a competitive assay, one treatment chemical is compared simultaneously to another to provide 
information on the relative attraction of one treatment to another. It also can provide information on the 
interfering effects from an inhibitor released in the opposite port and provide an indication about whether 
the inhibitor functions best when released at close range to the attractants, or if it can be released from 
another location and still be effective. The advantage of this technique is that it may provide a closer 
approximation to field situations where attractants or inhibitors must function in a complex situation 
against mosquitoes in competition with many other odors. Olfactometers that are used to assess 
the biological activity of candidate attractants have allowed the development of the human odor 
blends , 68 ’ 84 - 86 ' 101 - 102 ’ 111436 ’ 137 such as L-lactic acid and carbon dioxide , 138 L-lactic acid and ammonia , 111 
L-lactic acid and specific carboxylic acids , 86 and a three-component blend of L-lactic acid, acetone, and 
dimethyl disulfide . 84 ’ 85 ’ 102 ’ 137 

The development and use of a standard that has high attraction efficiency, reproducibility, and 
stability is important when conducting experiments to identify attraction-inhibitors. The use of such a 
blend has applicability to in vitro repellent experiments by obviating the need for volunteers to 
participate in in vivo studies. A standard chemical blend of attractants removes the variability inherent 
in the use of live hosts. Not only do individuals vary in their attractiveness and compound abundances 
detected, but a single human can vary substantially in both biological activity and compound 
abundances in their profile from day-to-day. 69 However, caution must be exercised in the interpretation 
of results from trials in which blends of attractant chemicals are used because they represent an 
approximation of a host. These mixtures consist of only a small number of kairomones and it is 
reasonably certain that of the hundreds of compounds emanated from human skin and some of the 
important attractants still remain unidentified. Most humans and skin extracts are still more attractive 
than our best synthetic blends when tested competitively in laboratory bioassays. 136 ’ 137 One of our 
bioassay protocols for attraction-inhibition involves comparing the response of a standard blend to the 
response of the same blend, delivered at the same dose but with a candidate attraction-inhibitor added to 
it. In other cases, the response of the candidate plus another known attractant like L-lactic acid is 
compared to the response to L-lactic acid alone when looking for synergism. Again, this method of 
testing attraction-inhibition may be even further removed from reality than using human odors or a more 
complex blend with higher attractiveness because as noted above, the human odor profile is significantly 
much more complex. 

Correlating Small- and Large-Scale Laboratory Results to Field Experiments 

One concern with results from laboratory bioassays is that they may not correlate well to the performance 
in the field. Laboratory bioassays are conducted under well-controlled conditions with the temperature, 
humidity, wind speed, and other variables controlled as needed. Although bioassays can involve 
movement in space, this movement is often confined. At best, the movement is in essence two 
dimensional, if not actually closer to a one-dimensional situation in which the mosquito travels linearly 
upwind through a tube. Additionally, bioassays in the laboratory may only examine a subset of all factors 
involved in host location, even though this may be intended partly by design. Laboratory olfactometers 
have a finite length or depth, and thus can best assess only the medium- to close-range stimuli. Finally, 
bioassays of this nature are considered to be undiscriminating assays in the treatise of Kennedy because 
the overall result, e.g., attraction, is analyzed as a complex of responses, rather than the individual 
isolated responses, as would be done in a discriminating assay. 21 


2006 by Taylor & Francis Group, LLC 



92 


Insect Repellents: Principles , Methods, and Uses 


Field Tests and Use of Spatial Repellents and Attraction-Inhibitors 
Experimental Design of Field Tests 

Use of Large-Cage Experiments and Laboratory-Reared Colony Mosquitoes 

Researchers have conducted large cage (9.1 m wideX 18.3 m longX4.9 m high, gabled to 5.5 m) studies 
to simulate environmental conditions that might be encountered in field studies against natural 
populations of mosquitoes. Traps releasing known attractants at specified release rates are placed in 
the center of the cage. 139 We choose to test with a 2.4 mX2.4 m designed perimeter around the trap. A 
wooden stake with an attached attraction-inhibitor releasing device is located at each corner of the 
perimeter. An inhibitor release device is attached 0.6 m above ground level to each stake. Both the 
inhibitor release device and trap are activated at least 30 min before mosquitoes were released into 
the cage, and operated for a specified time period, typically 12 h. At the conclusion of the test period, the 
trap collection device is retrieved and landing rates on humans are conducted within the cage at several 
established locations outside the 2.4 mX 2.4 m perimeter. The landing rate counts are performed in 
addition to trap collections to provide a more comprehensive indication of the effectiveness of the 
candidate attraction-inhibitor being tested. The benefits to using a large cage, similar to the benefits of 
laboratory studies, is that they provide a controlled setting with mosquitoes of known species 
composition, physiological and chronological age, and quantity. Furthermore, the escape of mosquitoes 
is minimized. However, the environmental conditions inside the cage are similar to those outside, as is 
the landscaping within the space. The drawback is that the mosquitoes are not allowed to migrate beyond 
the enclosure, as they would be able to do in the wild. 

Experiments with Wild Mosquitoes in the Field 

One concern with using colony-reared mosquitoes is whether or not they will behave similarly to those in 
the wild. Additionally, there are a variety of mosquito species and this composition can vary significantly 
during the course of a study. Conduction of field tests against natural populations of mosquitoes is 
performed in a similar experimental setup as that used in the large cage studies. A series of 2.4 mX 2.4 m 
plots can be established with traps, similarly baited as in the large cages, located in the center with the 
inhibitor dispensing devices placed on the four corners. A Latin square design can be used with days as 
replicates. 140 Initially, treatments and controls should be randomly assigned to each plot. The plots 
should be located far enough apart to prevent interactions among treatments. The treatments are then 
moved to new stations each day until all treatments have been evaluated in each plot at least once. Jensen 
et al. has used a variation of this design to evaluate citronella candles in Illinois. 141 At each sampling 
station in their study, the candles were arranged into an equilateral triangle, 3 m apart, with an individual 
measuring efficacy sitting in the center, about 1.5 m from each candle. The individual aspirated 
mosquitoes trying to bite exposed legs during four 15-min collection periods using a 
mechanical aspirator. 

Another study conducted by Lindsay et al. 142 in Canada used eight sampling stations arranged in a grid 
separated by at least 10 m. Two of each kind of dispenser were placed at each sampling period on top of 
35-cm-high plastic stands 1 m apart. A plastic lawn chair was placed between the plastic stands and 
subjects conducted biting counts while seated on the lawn chairs. The subjects were assigned to one of 
the eight sampling stations at the beginning of each evening and then rotated through all eight positions 
twice each night. Treatments were assigned to positions on the grid such that each treatment was at each 
position during the eight-night evaluation. It is important to evaluate each candidate product under a wide 
range of field conditions against a diversity of mosquito species, comparing their effectiveness to both 
negative (untreated) and positive (deet-treated individual) controls. Recently, Webb et al. used carbon 
dioxide-baited light traps and dispersed candidate inhibitors in a 4 unit X 4 unit grid with each of the 


© 2006 by Taylor & Francis Group, LLC 



Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


93 


16 dispensers about 1.5 m above ground. 143 Significant repellency was noted for catnip oil, deet, and the 
E,Z-dihydronepetalactone isomer from catnip oil. 

Use of Stand-Alone Inhibitor-Delivery Technology 

Currently, there is a commercial device that is on the market using inhibitor technology based on linalool. 
The active ingredient is primarily the (S)-( + )-linalool isomer [as opposed to (R)-( —[-linalool] in 
candles and sold under the trade name Conceal®. However, not all chemicals may be amenable to 
delivery by candle, and therefore devices similar to another commercial device, the Mosquito Cognito®,* 
may be an alternative approach to disperse low levels of inhibitor aerosols into the environment. The 
active ingredient is contained in cartridges and is used in a battery-powered device. 


Potential Applications of Spatial Repellents and Attraction-Inhibitors 
Species-Specific or Species-Exclusive Trapping 

At the present time, not enough is known about the concentration-dependent effects of attractants and 
attraction-inhibitors and how these chemicals may work or not work on many different mosquito species. 
Some inhibitors negatively affect the female mosquito at all concentrations tested, and against all species 
we have tested in the laboratory ( Anopheles quadrimaculatus. Anopheles abimanus, Aedes aegypti, 
Aedes albopictus, and Culex nigripalpus). 1 8 The rationale behind species-exclusive trapping would 
likely involve the use of odors based on avian emanations to selectively lure ornithophilic species of 
mosquitoes away from opportunistic feeding on a lesser-preferred host, such as humans. There is some 
basis for exploring this avenue of research because it has been shown that high (and/or low) levels of 
L-lactic acid are repellent for some species of mosquitoes, 78 ' 138 ' 144 and that specific species exhibit a 
strong host-preference based on emanated odors. 145 

Local Area Host-Finding Reduction 

One application of inhibitors has already been discussed in Section “Use of Stand-Alone Inhibitor- 
Delivery Technology”, i.e., the Mosquito Cognito/Conceal technology. The range of reduction in host 
finding is 50-95% with an average of 65% reduction based on tests in Sarasota, Vero Beach, and 
Loxahatchee (candles) and Lower Suwannee (candles) wildlife refuges in Florida. 146 It is possible that 
additional reduction might be achieved with the discovery of additional attraction-inhibitors. Also, it may 
be possible to design blends of inhibitors that may function synergistically in their effect, similar to that 
observed for chemicals used in kairomone blends that are derived from human odorants. 

Local Control Using a Push-Pull Strategy with Attractant-Baited Surveillance Traps 

Perhaps one of the greatest benefits to the development of potent inhibitors is the use of these compounds 
at a slow release rate to conceal host attractive odors in conjunction with surveillance traps to lure and 
trap or kill as a means of a barrier-forming push-pull strategy. 147 There are isolated situations, such as 
was shown in the work of Kline on Atsena Otie Key in Florida, where a reduction in mosquito biting 
incidence can be obtained using traps with attractants only. 148 This success is not expected to be possible 
in an area where competing host odors are constantly present. However, it is believed that even if there is 
a trap containing an attractant lure that is inferior to host odors, a push-pull strategy may overcome this 
and allow for local control in small areas. 


Registered trademark of BioSensory, Inc., Putnam, CT, USA. 


© 2006 by Taylor & Francis Group, LLC 





94 


Insect Repellents: Principles, Methods, and Uses 


Use of Structure-Activity Relationships to Benefit Development of Attraction-Inhibitors 

Scientists are exploring the use of quantitative structure-activity relationships (QSAR) as a means to 
examine repellents and to discover the structural basis that results in their biological activity. 149 ' 150 
Furthermore, this approach can be used as a means to predict novel molecular structures that are likely to be 
repellent. As attraction-inhibition becomes a more precisely characterized phenomenon, with increased 
numbers of inhibitors, dose response studies, and experiments designed to accurately assess inhibition 
level, these data should be amenable to QSAR studies. Through QSAR, researchers may also be able to 
predict the molecular and electronic properties of chemicals that result in attraction-inhibition. A 
comprehensive understanding of the chemicals could, in time, lead to a better understanding of the 
function of the odorant receptors. Extensive coverage of approaches to modeling repellents is found in 
Chapter 10 by Gupta and Bhattacharjee. 

References 

1. F. Flerbert, October 8 , 1920-February 11, 1986. 

2. H. H. Shorey, Interaction of insects with their chemical environment, in Chemical Control of Insect 
Behavior: Theory and Application, H. H. Shorey and J. J. McKelvey (Eds.), New York: Wiley, 1977, 
pp. 1-5. 

3. P. E. Howse, O. T. Jones, and I. D. R. Stevens, Pheromones and Behavior in Insect Pheromones and 
their Use in Pest Management, London: Chapman & Hall, 1998, pp. 3-6. 

4. V. G. Dethier, L. Barton Browne, and C. N. Smith, The designation of chemicals in terms of the 
responses they elicit from insects, J. Econ. EntomoL, 53, 134, 1960. 

5. P. Karlson and M. Liischer, “Pheromones”: A new term for a class of biologically active substances. 
Nature, 183, 55, 1959. 

6 . P. Karlson and M. Liischer, The proposed term “pheromone”. Nature, 183, 1835, 1959. 

7. C. I. Bliss, The toxicity of poisons applied jointly, Ann. Appl. Biol., 585, 26, 1939. 

8 . R. L. Plackett and P. S. Hewlett, Quantal responses to mixtures of poisons, J. Royal. Stat. Soc., 14B, 
141, 1952. 

9. R. H. Wright, D. L. Chambers, and I. Reiser, Insect attractants, anti-attractants, and repellents, Can. 
EntomoL, 103, 627, 1971. 

10. L. B. Browne, Host-related responses and their suppression: Some behavioral considerations, in 
Chemical Control of Insect Behavior: Theoiy and Application, H. H. Shorey and 
J. J. McKelvey (Eds.), New York: Wiley, 1977, pp. 117-127. 

11. H. K. Gouck, T. P. McGovern, and M. Beroza, Chemicals tested as space repellents against yellow- 
fever mosquitoes, I. Esters, J. Econ. EntomoL, 60, 1587, 1967. 

12. J. A. Nolen et al.. Method, Apparatus, and Compositions for Inhibiting the Human Scent Tracking 
Ability of Mosquitoes in Environmentally Defined Three Dimensional spaces, Washington, D.C.: U.S. 
Patent and Trademark Office, 2002 (U.S. Patent No. 6,362,365). 

13. C. E. Schreck and D. L. Kline, Area protection by use of repellent-treated netting against Culicoides 
biting midges, Mosq. News, 43, 338, 1983. 

14. R. A. Wirtz, J. D. Turrentine, Jr., and L. C. Rutledge, Mosquito area repellents: Laboratory testing of 
candidate materials against Aedes aegypti, Mosq. News, 40, 432, 1980. 

15. H. Kawada et al.. Field evaluation of spatial repellency of metofluthrin impregnated plastic strips 
against mosquitoes in Hai Phong City, Vietnam, Am. J. Trop. Med. Hyg., 73, 350, 2005. 

16. D. R. Roberts et al.. Methods of testing and analyzing excito-repellency responses of malaria vectors 
to insecticides, J. Am. Mosq. Control Assoc., 13, 13, 1997. 

17. K. Ujihara et al., Metofluthrin: A potent new synthetic pyrethroid with high vapor activity against 
mosquitoes, Biosci. Biotechnol. Biochem., 68 , 170, 2004. 

18. M. J. Klowden, The physiological control of mosquito host-seeking behavior, in Current Topics in 
Vector Research, K. F. Harris (Ed.), Vol. 1, Westport, Connecticut: Praeger Scientific, 1983, 
pp. 93-116. 


© 2006 by Taylor & Francis Group, LLC 


Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


95 


19. J. E. Simpson and R. H. Wright, Area treatment to combat mosquitoes, Nature , 214, 113, 1967. 

20. W. L. Roelofs and A. Comeau, Sex pheromone perception: Synergists and inhibitors for the 
red-banded leaf roller attractant, J. Insect Physiol., 17, 435, 1971. 

21. J. S. Kennedy, Behaviorally discriminating assays of attractants and repellents, in Chemical Control of 
Insect Behavior: Theory and Application, H. H. Shorey and J. J. McKelvey (Eds.), New York: Wiley, 
1977, pp. 215-229. 

22. E. E. Davis, Structure-response relationship of the lactic-acid excited neurons in the antennal grooved- 
peg sensilla of the mosquito Aedes aegypti, J. Insect Physiol., 34, 443, 1988. 

23. S. J. Torr, T. N. C. Mangwiro, and D. R. Hall, Responses of Glossina Pallidipes (Diptera: Glossinidae) 
to synthetic repellents in the field, Bull. Entomol Res., 86, 609, 1996. 

24. E. B. Dogan and P. A. Rossignol, An olfactometer for discriminating between attraction, inhibition, 
and repellency in mosquitoes (Diptera: Culicidae), J. Med. Entomol, 36, 788, 1999. 

25. E. B. Dogan, J. W. Ayres, and P. A. Rossignol, Behavioural mode of action of deet: Inhibition of lactic 
acid attraction, Med. Vet. Entomol., 13, 97, 1999. 

26. W. A. Skinner et al., Repellency of skin-surface lipids of humans to mosquitoes. Science, 149, 305, 
1965. 

27. H. I. Maibach et al., Factors that attract and repel mosquitoes in human skin, J. Am. Med. Assoc.. 196, 
173, 1966. 

28. H. I. Maibach et al.. Human skin in relationship to mosquito attraction and repulsion, Conn. Med., 33,23, 

1969. 

29. W. A. Skinner et al.. Human skin-surface lipid fatty acid—Mosquito repellents, Experientia, 26, 728, 

1970. 

30. S. R. Christophers, Mosquito repellents. Being a report of the work of the mosquito repellent inquiry, 
Cambridge (1943-5), J. Hygiene., 45, 176, 1947. 

31. R. N. McCulloch and D. F. Waterhouse, Laboratory and field tests of mosquito repellents. Bull. Council. 
Sci. Indust. Res. Aust., 213, 28, 1947. 

32. USDA, Report of Work Conducted Under Funds Allotted by the Department of National Defense, 
Fourth Quarterly Progress Report, Orlando, FL, 1948. 

33. T. P. McGovern, M. Beroza, and H. Gouck, Chemicals tested as space repellents against yellow-fever 
mosquitoes, II. Carbanilates, benzamides, aliphatic amides, and imides, J. Econ. Entomol., 60, 1591, 
1967. 

34. C. E. Schreck et al., Spatial action of mosquito repellents, J. Econ. Entomol, 63, 1576, 1970. 

35. A. A. Khan and H. I. Maibach, A study of insect repellents, 1. Effect on flight and approach by Aedes 
aegypti, J. Econ. Entomol., 65, 1318, 1972. 

36. T. Eisner, Catnip: Its raison d’etre. Science, 146, 1318, 1964. 

37. L. R. Garson and M. E. Winnike, Relationships between insect repellency and chemical and physical 
parameters—A review, J. Med. Entomol, 5, 339, 1968. 

38. Peterson, C. J. Insect repellents of natural origin: Catnip and osage orange, Ph.D. dissertation, Iowa 
State University, Ames, IA, 2001. 

39. C. Peterson and J. Coats, Insect repellents—past, present, and future, Pestic. Outlook., 12, 154, 2001. 

40. U. R. Bernier et al.. Comparison of contact and spatial repellency of catnip oil and AQV-diethyl-3- 
methylbenzamide (deet) against mosquitoes, J. Med. Entomol., 42, 306, 2005. 

41. K. R. Chauhan et al.. Feeding deterrent effects of catnip oil components compared with two synthetic 
amides against Aedes aegypti, J. Med. Entomol., 42, 643, 2005. 

42. J. Boeckh et al., Acylated 1,3-aminopropanols as repellents against bloodsucking arthropods, Pestic. 
Sc;., 48, 359, 1996. 

43. E. J. Hoffman and J. R. Miller, Reduction of mosquito (Diptera: Culicidae) attacks on a human subject by 
combination of wind and vapor-phase deet repellent, J. Med. Entomol, 39, 935, 2002. 

44. U. R. Bernier, Unpublished data, 1997. 

45. Z. A. Mehr et al.. Attraction of mosquitoes to diethyl methylbenzamide and ethyl hexanediol, J. Am. 
Mosq. Control Assoc., 6, 469, 1990. 

46. K. H. Posey, D. R. Barnard, and C. E. Schreck, Triple cage olfactometer for evaluating mosquito 
(Diptera: Culicidae) attraction responses, J. Med. Entomol., 35, 330, 1998. 


2006 by Taylor & Francis Group, LLC 



96 


Insect Repellents: Principles, Methods, and Uses 


47. E. J. Hoffman and J. R. Miller, Reassessment of the role and utility of wine in suppression of mosquito 
(Diptera: Culicidae) host finding: Stimulus dilution supported over flight limitation, J. Med. Entomol., 
40, 607, 2003. 

48. R. R. Painter, Repellents, in Pest Control: Biological, Physical and Selected Chemical Methods, 
W. W. Kilgore and R. R. Painter (Eds.), New York: Academic Press, 1967. 

49. G. F. Shambaugh, R. F. Brown, and J. J. Pratt, Jr., Repellents for biting arthropods, in Advances 
in Pest Control Research, Vol. 1, R. L. Metcalf (Ed.), New York: Interscience Publishers, 1957, 
pp. 277-303. 

50. K. Narusuye et al., Linalool suppresses voltage-gated currents in sensory neurons and cerebellar 
purkinje cells, J. Neural Trans., 112, 193, 2005. 

51. M. A. Birkett et al.. The role of volatile semiochemicals in mediating host location and selection 
by nuisance and disease-transmitting cattle flies, Med. Vet. Entomol., 18, 313, 2004. 

52. R. H. Bedoukian, Method and Compositions for Inhibiting the Scent Tracking Ability of 
Mosquitoes and Biting Midges, Washington D.C.: U.S. Patent and Trademark Office, 2005. 
Published U.S. Patent Application No. 20050090563. 

53. D. L. Kline et al., Olfactometric evaluation of spatial repellents for Aedes aegypti, J. Med. 
Entomol., 40, 463, 2003. 

54. M. Markowitz, Practical Survey of Chemistry and Metabolism of the Skin, Philadelphia: Blakiston 
Company, 1942. 

55. H. C. Koring et al., Changes in skin pH and resident flora by washing and synthetic detergent 
preparations at pH 5.5 and 8.5, J. Soc. Cosmet. Chem., 42, 147, 1991. 

56. N. Nicolaides, H. C. Fu, and G. R. Rice, The skin surface lipids of man compared with those of 
eighteen species of animals, J. Invest. Dermatol., 51, 83, 1968. 

57. N. Nicolaides, H. C. Fu. and M. N. A. Ansari, Diester waxes in surface lipids of animal skin, 
Lipids, 5, 299-307, 1970. 

58. N. Nicolaides et al.. Lipid composition of comedomes compared with that of human skin surface 
in acne patients, J. Invest. Dermatol., 54, 487, 1970. 

59. M. N. A. Ansari, N. Nicolaides, and H. C. Fu, Fatty acid composition of the living layer and 
stratum corneum lipids of human sole skin epidermis. Lipids, 5, 838, 1970. 

60. S. D. Sastry et al.. Volatiles emitted by humans, in Biochemical Applications of Mass 
Spectrometery, First Supplementary Volume, G. R. Waller and O. C. Dermer et al. (Eds.), 
Chichester: Wiley, 1980, pp. 1085-1129. 

61. R. I. Ellin et al.. An apparatus for the detection and quantitation of volatile human effluents, 
J. Chromatogr., 100, 137, 1974. 

62. K. Naitoh et al., New measurement of hydrogen gas and acetone vapor in gases emanating from 
human skin, Instr. Sci. Technoi, 30, 267, 2002. 

63. Y. Shirai et al.. Alcohol ingestion stimulates mosquito attraction, J. Am. Mosq. Control Assoc., 
18, 91, 2002. 

64. U. R. Bernier, M. M. Booth, and R. A. Yost, Analysis of human skin emanations by gas 
chromatography/mass spectrometry, 1. Thermal desorption of attractants for the yellow fever 
mosquito ( Aedes aegypti ) from handled glass beads, Anal. Chem., 71, 1, 1999. 

65. X.-N. Zeng et al.. Analysis of characteristic odors from human male axillae, J. Chem. Ecol., 17, 
1469, 1991. 

66 . X.-N. Zeng et al.. An investigation of human apocrine gland secretion for axillary odor 
precursors, J. Chem. Ecol., 18, 1039, 1992. 

67. U. R. Bernier et al.. Analysis of human skin emanations by gas chromatography/mass 
spectrometry, 2. Identification of volatile compounds that are candidate attractants for the 
yellow fever mosquito ( Aedes aegypti), Anal. Chem., 72, 747, 2000. 

68 . U. R. Bernier et al., Chemical analysis of human skin emantions: Comparison of volatiles from 
humans that differ in attraction of Aedes aegypti (Diptera: Culicidae), J. Am. Mosq. Control 
Assoc., 18, 186, 2002. 

69. C. E. Schreck, D. L. Kline, and D. A. Carlson, Mosquito attraction to substances from the skin of 
different humans, J. Am. Mosq. Control Assoc., 6, 406, 1990. 


© 2006 by Taylor & Francis Group, LLC 



Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


97 


70. K. G. Asano et al.. Chemical composition of fingerprints for gender determination, J. Forensic 
Sci., 47, 805, 2002. 

71. A. M. Curran, S. I. Rabin, and K. G. Furton. Analysis of the uniqueness and persistence of 
human scent, Forensic Sci. Commun., 72, 1, 2005. 

72. A. M. Curran et al.. Comparison of the volatile organic compounds present in human odor using 
spme-gc/ms, J. Chem. Ecol., 31, 1607, 2005. 

73. W. Ma and W. R. Klemm, Determination of acetaldehyde in blood by solid phase extraction and 
high performance liquid chromatography, Alcohol, 14, 469, 1997. 

74. R. Andreoli et al.. Determination of patterns of biologically relevant aldehydes in exhaled breath 
condensate of healthy subjects by liquid chromatography/atmospheric chemical ionization tandem 
mass spectrometry, Rapid Commun. Mass Spectrom., 17, 637, 2003. 

75. B. Krotoszynski, G. Gabriel, and H. O’Neill, Characterization of human expired air: A promising 
investigative and diagnostic technique, J. Chromatogr. Sci., 15, 239-244, 1977. 

76. M. Phillips, Method for the collection and assay of volatile organic compounds in breath, Anal. 
Biochem., 247, 272, 1997. 

77. D. L. Ashley et al.. Determining volatile organic compounds in human blood from a large sample 
population by using purge and trap gas chromatography/mass spectrometry, Anal. Chem., 64, 
1021, 1992. 

78. W. Rudolfs, Chemotropism of mosquitoes. Bull. N.J. Agric. Exp. Sta., 367, 4, 1922. 

79. M. Bar-Zeev, H. I. Maibach, and A. A. Khan, Studies on the attraction of Aedes aegypti (Diptera: 
Culicidae) to man, J. Med. Entomol., 14, 113, 1977. 

80. M. T. Gillies, The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae): A 
review. Bull. Ent. Res., 70, 525, 1980. 

81. W. Takken, The role of olfaction in host-seeking of mosquitoes: A review, Insect Sci. Applic., 12, 
287, 1991. 

82. T. Dekker, M. Geier, and R. T. Carde, Carbon dioxide instantly sensitizes female yellow fever 
mosquitoes to human skin odors, J. Exp. Biol, 208, 2963, 2005. 

83. W. R. Mukabana et al., Allomonal effect of breath contributes to differential attractiveness of 
humans to the African malaria vector Anopheles gambiae, Malar. J., 3, 1, 2004. (http://www. 
malariajournal.com/content/3/1/1). 

84. U. R. Bernier et al.. Chemical Composition that Attract Arthropods, Washington D.C.: U.S. 
Patent and Trademark Office, 2001, U.S. Patent No. 6,267,953. 

85. U. R. Bernier et al.. Chemical Composition that Attract Arthropods, Washington D.C.: U.S. 
Patent and Trademark Office, 2004, U.S. Patent No. 6,800,279. 

86 . O. J. Bosch, M. Geier, and J. Boeckh, Contribution of fatty acids to olfactory host finding of 
female Aedes aegypti, Chem. Senses., 25, 323, 2000. 

87. R. C. Smallegange et al.. Synergism between ammonia, lactic acid and carboxylic acids as 
kairomones in the host-seeking behavior of the malaria mosquito Anopheles gambiae sensu stricto 
(Diptera: Culicidae), Chem. Senses., 30, 145, 2005. 

88 . C. Constantini et al., Electroantennogram and behavioral response of the malaria vector 
Anopheles gambiae to human-specific sweat components, Med. Vet. Entomol., 15, 259, 2001. 

89. W. G. Reifenrath, Natural Insect Repellant, Washington D.C.: U.S. Patent and Trademark Office, 
2005, U.S. Patent No. 6,953,814. 

90. H. D. Douglas III, et al., Heteropteran chemical repellents identified in the citrus odor of a 
seabird (crested auklet: Aethia cristatella): Evolutionary convergence in chemical ecology, 
Naturwissenschaften, 88, 330, 2001. 

91. H. D. Douglas III, et al.. Interspecific differences in Aethia spp. auklet odorants and evidence for 
chemical defense against ectoparasites, J. Chem. Ecol., 30, 1921, 2001. 

92. H. D. Douglas III, et al., Chemical odorant of colonial seabird repels mosquitoes, J. Med. 
Entomol., 42, 647, 2005. 

93. J. C. Hagelin, I. L. Jones, and L.E L. Rasmussen, A tangerine-scented social odour in a 
monogamous seabird, Proc. R. Soc. Loud., 270B, 1323, 2003. 


© 2006 by Taylor & Francis Group, LLC 


98 


Insect Repellents: Principles, Methods, and Uses 


94. P. Steullet and P. M. Guerin, Identification of vertebrate volatiles stimulating olfactory receptors on 
tarsus I of the tick Amblyomma Variegatum Fabricus (Ixodidae), J. Comp. Physiol, 174A, 27, 
1994. 

95. P. G. Guerenstein and P. M. Guerin, Olfactory responses of the blood-sucking bug Triatoma 
infestans to odours of vertebrate hosts, J. Exp. Biol, 204, 585, 2001. 

96. V. G. Dethier, Olfactory responses of blowflies to aliphatic aldehydes, J. Gen. Physiol., 37, 743, 
1954. 

97. S. Haze et al., 2-Nonenal newly found in human body odor tends to increase with aging, J. Invest. 
Dermatol., 116, 520, 2001. 

98. Z.-M. Zhang et al.. The study of fingerprint characteristics of the emanations from human arm skin 
using the original sampling system by spme-gc/ms, J. Chromatogr., 822B, 244, 2005. 

99. Y. K. Takahashi, S. Nagayama, and K. Mori, Detection of spoiled food smells by odor maps in the 
olfactory bulb, J. Neurosci., 24, 8690, 2004. 

100. M. Corradi et al.. Aldehydes in exhaled breath condensate of patients with chronic obstructive 
pulmonary disease, Am. J. Respir. Crit. Care Med., 167, 1380, 2003. 

101. U. R. Bernier et al.. Unpublished data, 2005. 

102. U. R. Bernier et al.. Synergistic attraction of Aedes aegypti (L.) to binary blends of L-lactic acid and 
acetone, dichloromethane, or dimethyl disulfide, J. Med. Entomol., 40, 653, 2003. 

103. A. W. Ralston and J. P. Barrett, Insect Repellent Containing Aliphatic Alcohols, Washington D.C.: 
U.S. Patent and Trademark Office, 1941, U.S. Patent No. 2,254,665. 

104. B. G. Wilkes, Insect Repellents, Washington D.C.: U.S. Patent and Trademark Office, 1946, U.S. 
Patent No. 2,407,225. 

105. P. Granett and H. L. Haynes, Insect-repellent properties of 2-ethylhexanediol-l,3, J. Econ. 
Entomol., 38, 671, 1945. 

106. D. R. Hall et al., l-Octen-3-ol: A potent olfactory stimulant and attractant for tsetse isolated from 
cattle odors. Insect Sci. Applic., 7, 5, 1984. 

107. W. Takken and D. L. Kline, Carbon dioxide and l-octen-3-ol as mosquito attractants, J. Amer. 
Mosq. Control Assoc., 5, 311, 1989. 

108. U. R. Bernier, Unpublished data, 2000. 

109. J. P. Conkle et al.. Detailed study of contaminant production in a space cabin simulator at 760 mm 
of mercury, Aerospace Med., 38, 491, 1967. 

110. P. A. Wright, Nitrogen excretion: Three end products, many physiological roles, J. Exper. Biol., 
198, 273, 1995. 

111. M. Geier, O. J. Bosch, and J. Boeckh, Ammonia as an attractive component of host odour for the 
yellow fever mosquito, Aedes aegypti, Chem. Senses., 24, 647, 1999. 

112. M. A. H. Braks, J. Meijerink, and W. Takken, The response of the malaria mosquito. Anopheles 
gambiae, to two components of human sweat, ammonia and L-lactic acid, in an olfactometer. 
Physiol. Entomol., 26, 142, 2001. 

113. U. R. Bernier, Unpublished data, 1999. 

114. C. W. O. Bunker and A. D. Hirshfelder, Mosquito repellents. Am. J. Trop. Med., 5, 359, 1925. 

115. L. A. O. Roadhouse, Laboratory studies on insect repellency. Can. J. Zool., 31, 535, 1953. 

116. V. G. Dethier, Repellents, Ann. Rev. Entomol., 1, 181, 1956. 

117. A. Cork and K. C. Park, Identification of electrophysiologically-active compounds for the 
malaria mosquito. Anopheles gambiae, in human sweat extracts, Med. Vet. Entomol, 10, 269, 
1996. 

118. A. Ostrovskaya et al.. Study and identification of volatile compounds from human skin, 
J. Cosmetic Sci., 53, 147, 2002. 

119. F. Kanda et al.. Elucidation of chemical compounds responsible for foot malodour, Brit. 
J. Dermatol., 122, 771, 1990. 

120. E. D. Adrian, The activity of the nervous system in the caterpillar, J. Physiol., 70, 34, 1930. 

121. W. Roelofs et al., Sex attractant of the colding moth: Characterization with electroantennogram 
technique. Science, 174, 291, 1971. 


© 2006 by Taylor & Francis Group, LLC 



Natural Compounds That Inhibit Mosquito Host-Finding Abilities 


99 


122. W. C. Agosta, Chemical Communication: The Language of Pheromones, Scientific American 
Library, New York: W.H. Freeman & Co., 1992, p. 66. 

123. A. Cork et al., Gas chromatography linked to electroantennography: A versatile technique 
for identifying insect semiochemicals, in Chromatography and Isolation of Insect Hormones and 
Pheromones, A. R. McCaffery, D. Wilson et al. (Eds.), London: Plenum Press, 1990. 

124. A. Cork. Olfactory basis of host-finding in haemotophagous arthropods, in Olfaction in 
Mosquito-Host Interactions, Ciba Foundation Symposium No. 200, G. R. Cardew (Ed.), 
London: John Wiley & Sons, 1996. 

125. J. Meijerink et al.. Identification of olfactory stimulants for Anopheles gambiae from human 
sweat samples, J. Chem. Ecol., 26, 1367, 2000. 

126. Y. T. Qiu et al.. Behavioural and electrophysiological responses of the malaria mosquito 
Anopheles gambiae Giles sensu stricto (Diptera: Culicidae) to human skin emanations, Med. 
Vet. Entomol., 18, 429, 2004. 

127. E. Bursell et al.. Identification of compounds of cattle urine attractive to tsetse flies. Glossina 
spp. (Diptera: Glossinidae), Bull. Entomol. Res., 78, 281, 1988. 

128. A. Cork, Identification of electrophyisologically-active compounds for new world screwworm, 
Cochliomyia hominovorax, in larval wound fluid, Med. Vet. Entomol., 8, 151, 1994. 

129. W. M. Hoskins and R. Craig, The olfactory responses of flies in a new type of insect 
olfactometer, I. Theory and design of the olfactometer, J. Econ. Entomol., 27, 1029, 1934. 

130. J. O. G. Wieting and W. M. Hoskins, The olfactory responses of flies in a new type of insect 
olfactometer, II. Responses of the housefly to ammonia, carbon dioxide and ethyl alcohol, 
J. Econ. Entomol., 32, 24, 1939. 

131. E. R. Willis, The olfactory responses of female mosquitoes, J. Econ. Entomol., 40, 769, 
1947. 

132. H. K. Gouck and C. E. Schreck, An olfactometer for use in the study of mosquito 
attractants, J. Econ. Entomol, 58, 589, 1965. 

133. C. E. Schreck, H. K. Gouck, and N. Smith, An improved olfactometer for use in studying 
mosquito attractants and repellents, J. Econ. Entomol, 60, 1188, 1967. 

134. J. P. Grieco et al., A novel high-throughput screening system to evaluate the behavioral 
response of adult mosquitoes to chemicals, J. Am. Mosq. Control Assoc., 21, 404, 2005. 

135. W. M. Barrows, The reactions of the pomace fly. Drosophila ampelophila Loew, to odorous 
substances, J. Exper. Zool, 4, 515, 1907. 

136. M. Geier, H. Sass, and J. Boeckh, A search for components in human body odour that 

attract females of Aedes aegypti, in Olfaction in Mosquito-Host Interactions, Ciba Foundation 

Symposium No. 200, G. R. Cardew (Ed.), London: John Wiley & Sons, 1996. 

137. U. R. Bernier et al., Laboratory comparison of Aedes aegypti (L.) attraction to human odors 

and to synthetic human odor compounds and blends, J. Am. Mosq. Control Assoc., in press. 

138. C. N. Smith et al., L-lactic acid as a factor in the attraction of Aedes aegypti (Diptera 

Culicidae) to human hosts, Ann. Entomol Soc. Am., 63, 760, 1970. 

139. D. L. Kline, Unpublished data, 2000. 

140. W. G. Cochran and G. M. Cox, Experimental Design, New York: Wiley, 1957. 

141. T. Jensen et al., Field efficacy of commercial antimosquito products in Illinois, J. Am. Mosq. 
Control Assoc., 16, 148, 2000. 

142. L. R. Lindsay et al.. Evaluation of the efficacy of 3% citronella candles and 5% citronella 
incense for protection against field populations of Aedes mosquitoes, J. Am. Mosq. Control 
Assoc., 12, 293, 1996. 

143. R. E. Webb et al., Evaluation of spatial repellents against mosquitoes in the field, J. Am. 
Mosq. Control Assoc., in press. 

144. Y. Shirai et al., L-lactic acid as a mosquito (Diptera: Culicidae) repellent on human and 
mouse skin, J. Med. Entomol, 38, 51, 2001. 

145. S. A. Allan, U. R. Bernier, and D. L. Kline, Laboratory evaluation of avian odors for 
mosquito (Diptera: Culicidae) attraction, J. Med. Entomol., in press. 

146. J. A. Nolen, Personal communication, 2006. 


2006 by Taylor & Francis Group, LLC 



100 


Insect Repellents: Principles , Methods, and Uses 


147. J. R. Miller and R. S. Cowles, Stimulo-deterrent diversion: A concept and its possible 
application to onion maggot control, J. Chem. Ecol., 16, 3197, 1990. 

148. D. L. Kline, Unpublished data, 2005. 

149. C. Hansch, A. Leo, and D. H. Hoekman, Exploring QSAR., ACS Professional Reference 
Book , Washington D.C.: American Chemical Society, 1995. 

150. D. Ma et al.. Predicting mosquito repellent potency of AL/V-diethyl-m-toluamide (deet) analogs 
from molecular electronic properties, Am. J. Trop. Med. Hygiene., 60, 1, 1999. 


© 2006 by Taylor & Francis Group, LLC 



5 


Standard Methods for Testing Mosquito Repellents 


Donald R. Barnard, Ulrich R. Bernier, Rui-de Xue, and Mustapha Debboun 


CONTENTS 

Introduction.103 

Laboratory Repellent Bioassay Methods.104 

World Health Organization Method.104 

American Society for Testing and Materials Method E951-94 (Revised 2000).105 

Screened Cage Method.105 

K&D Module Method.106 

Field Repellent Bioassay Methods.107 

World Health Organization Method.107 

American Society for Testing and Materials Method E939-94 (Revised 2000).107 

U.S. Environmental Protection Agency Test Guidelines.107 

Sources of Variation in Repellent Bioassays.108 

Abiotic Factors.108 

Biotic Factors.108 

Conclusions.109 

References.109 


Introduction 

Testing of mosquito repellents, whether in the laboratory or the field, is performed using a process called 
biological assay (bioassay for short). 1 Bioassays can be used to answer three questions about repellents: 

1. Is the candidate material repellent? 

2. What quantity of material is required for repellency? 

3. How long does repellency last? 

There are three repellent bioassay procedures documented as standard methods in the literature and 
one set of repellent testing guidelines available on the internet: 

1. World Health Organization (WHO), WHO/Control of Tropical DiseaseAVHO Pesticide 
Evaluation Scheme/Informal Consultation (WHO/CTD/WHOPES/IC 96.1). Report of 
WHOPES Informal Consultation on the Evaluation and Testing of Insecticides. 2 


103 


© 2006 by Taylor & Francis Group, LLC 






















104 


Insect Repellents: Principles , Methods, and Uses 


2. American Society for Testing and Materials (ASTM) E951-94 (revised 2000). Laboratory 
Testing of Non-Commercial Repellant Formulations on the Skin. 3 

3. American Society for Testing and Materials ( ASTM) E939-94 (revised 2000). Field Testing 
Topical Applications of Compounds as Repellents for Medically Important and Pest 
Arthropods. 1. Mosquitoes. 4 5 

4. U.S. Environmental Protection Agency (EPA) Office of Prevention, Pesticides, and Toxic 
Substances (OPPTS) 810.3700. Product Performance Test Guidelines. Insect Repellents for 
ffuman Skin and Outdoor Premises. EPA #712-C-99-369, December 1999 (available in 
public draft on the internet; see below for URL). 

A fifth repellent bioassay system, the screened cage method, is frequently cited in the literature, 5-8 and 
a sixth system, modified from ASTM E951-94 and adaptable to both in vivo and in vitro testing, has 
recently been published. 9-11 


Laboratory Repellent Bioassay Methods 
World Health Organization Method 

Laboratory repellent bioassays based on the WHO protocol 1 2 require a mosquito-filled, screened cage and 
use deet as a positive control. Human test subjects are preferred over laboratory animals or artificial 
membranes. Aedes aegypti, the normal test species, is used in variable numbers, but other mosquito 
species can be substituted depending on the needs of the experiment. An area of skin ranging from that 
covering the entire forearm to as little as 25 cm 2 is treated with repellent and exposed to caged 
mosquitoes. Untreated skin is covered with a glove or other protective material. For compounds of 
unknown toxicology, the repellent may be applied to a cotton stockinette sleeve, and the treated sleeve 
may be pulled over a second untreated stockinette on the arm to prevent skin contact with the repellent. 

At least five variations of the WHO method have been developed to meet the testing needs of different 
institutions. 2 These meathods emphasize either the determination of protection time after treatment with 
a single repellent dose or the percent protection in relation to repellent dose. The protocols are as follow: 

1. A 25 cm 2 area on a subject’s forearm is treated with an ethanolic solution of repellent 
(treatment), and the same-sized area on the adjacent forearm is treated with alcohol (negative 
control). Both arms are simultaneously introduced into one cage, and the numbers of 
mosquitoes biting each arm in 5 min is recorded. Percent protection is calculated by 
comparing biting rates on the treatment and control arms. 

2. A subject’s feet and legs are treated with repellent, exposed to 25 female mosquitoes in a 
mosquito-proof enclosure (lmXlmX3m high), and the number of bites in 10 min 
is recorded. 

3. A subject’s forearm is treated with 1 mL of a 25% ethanolic repellent solution and introduced 
into a mosquito-filled cage for 3 min once every 30 min. Protection time is that elapsed 
between repellent application and the first mosquito bites followed by a confirmatory bite in 
the same, or next, exposure period. 

4. One gram of repellent is dissolved in sufficient acetone to saturate 280 cm 2 of cotton 
stockinette. The stockinette is drawn over the arm of a subject and exposed to 1,500 caged 
female mosquitoes for 1 min, at daily or weekly intervals, until 5 bites are obtained. 

5. A subject’s untreated arm is exposed to 50 caged female mosquitoes, followed by repeated 
exposures of the same arm with increasingly high doses of repellent. In each exposure, the 


© 2006 by Taylor & Francis Group, LLC 



Standard Methods for Testing Mosquito Repellents 


105 


arm is withdrawn before the mosquitoes can imbibe blood. Probit analysis is used to calculate 
the ED 50 . When the dose giving 100% repellency is identified, the arm is re-exposed at 60 min 
intervals until repellency declines to 50%. 

American Society for Testing and Materials Method E951-94 (Revised 2000) 

This method 3 comprises the use of a rectangular (18 cm length X 5 cm width X 4 cm height) clear plastic 
test cage with five 29-mm-diameter openings in the bottom. A template is used to place four repellent 
dosages and a control on the skin of a human volunteer in a pattern that matches the openings on the test 
cage bottom. The cage is strapped to the arm or leg of a volunteer, bottom-side to the skin, with 10-15 
nulliparous, 5-15-day-old female mosquitoes placed into the cage through a 13-mm opening at one end. 
A test commences when the plastic slide (0.3-mm thick) that blocks the openings in the test cage bottom 
is withdrawn, allowing mosquitoes access to the repellent treated skin. The number of mosquitoes that 
land on and probe the skin in 2.5 min is recorded. The dose-response data obtained with ASTM E951-94 
has been used to calculate median (ED 50 ) and 95% effective doses (ED 95 ) 12 ’ 13 and to describe functional 
responses, in time, of mosquitoes to topical repellents. 12 

Screened Cage Method 

The screened cage bioassay method employs a 40 cm 3 aluminum-frame cage with a metal bottom, 
screened top and back, clear acrylic sides (for viewing), and a front stockinette sleeve for access. Two 
hundred human host-seeking 14 nulliparous, 7-8-day-old female mosquitoes are placed in the cage 1 h 
before the test. Treatment consists of a 25% ethanolic solution of repellent active ingredient applied to 
the forearm of a volunteer (between the wrist and elbow) at the rate of 1 mL/650 cm 2 of skin surface area. 
The treated forearm is inserted into the cage (a glove is used to protect the hand from mosquito bites. 
Figure 5.1) and the number of mosquitoes that land and probe the skin in 3 min is observed and recorded. 
The observations are repeated every 30 or 60 min. Two bites in one 3 min test or one bite in one 3 min 
test, followed by one or more bites in a second test 30 min later ends the test for the repellent. A second 
cage of mosquitoes is used as a positive or negative control. Depending upon the requirements of the 
experiment, protection time is calculated as either the time elapsed between repellent application and the 



FIGURE 5.1 The screen cage method of testing trial repellent formulations. 


© 2006 by Taylor & Francis Group, LLC 


106 


Insect Repellents: Principles, Methods, and Uses 


first confirmed mosquito bite, or the time between repellent application and the observation period 
immediately preceding the first confirmed bite. Data obtained with this bioassay method can be used to 
calculate complete protection time (CPT). 

K&D Module Method 

An extension of ASTM E951-94 is the “K&D module”. 9 This apparatus is reported to minimize the 
likelihood of treatment interactions. It increases the number of possible treatments per replicate and 
permits large numbers of replicated observations for each human test subject. The module can be used to 
test the responses of more than one mosquito species at a time to one dose of repellent or to evaluate 
repellent responses in the same species using specimens from geographically distinct locations. 

In vivo bioassays using the K&D module (Figure 5.2) are conducted in a walk-in incubator (27°C and 
80% RH) under fluorescent light. A template is used to delineate 3 cmX4 cm areas on the skin that 
correspond to each of the six cell openings on the bottom of the module. A treatment is administered by 
pipette onto a 4 cm X 5 cm rectangular area of skin centered over one of the individual template marks on 
a human volunteer and consists of 55 pL of ethanol containing 8.73 nmol of candidate repellent per 
microlitre of ethanol. This process results in a 24 nmol dose of the treatment on 1 cm 2 of skin. Skin 
treated with ethanol serves as the control. The module, with five mosquitoes in each cell, is then 
positioned over the treated skin area, each cell door is opened, and the number of mosquitoes that bite the 
skin or become blood-engorged in 2 min is recorded. 

The in vitro system 10 ' 11 (Figure 5.3) consists of six reservoirs (3 cmX4 cm) warmed to 38°C by a 
water bath. Each reservoir is filled with 6 mL of outdated human blood and covered with a Baudruche or 
collagen membrane. Trial repellent compounds dissolved in 110 pL of ethanol are applied in random 
order to six 4 cm X 5 cm pieces of organdy cloth. Each cloth is allowed to dry and then placed over one of 
the membrane-covered, blood-filled cells. The module, with five mosquitoes in each cell, is positioned 
over the treated cloth, and the doors are opened. The number of mosquitoes with their probosces inserted 
through the cloth into the Baudruche or collagen membrane into the blood after 2 min is recorded. 



FIGURE 5.2 The in vivo K&D module apparatus for bioassay of repellent active ingredients and formulations. 


© 2006 by Taylor & Francis Group, LLC 







Standard Methods for Testing Mosquito Repellents 


107 



FIGURE 5.3 The in vitro K&D module apparatus for bioassay of repellent active ingredients and formulations. 


Field Repellent Bioassay Methods 
World Health Organization Method 

When using the WHO field method, 2 repellent tests are made in the vicinity of human domiciles. 
Mosquito biting rate and the assessment of repellency is based on the capture of mosquitoes attacking 
human volunteers; thus, tests are timed to exploit the biting cycle of the target mosquito species. Test 
subjects are spaced 10 m apart and rotated in a randomized manner throughout the experiment to 
minimize positional errors. Appropriate criteria for repellency include 80% reduction in biting rate for 
6-8 h without adverse user side effects. 


American Society for Testing and Materials Method E939-94 (Revised 2000) 

In this method, 4 1.5 mL of repellent solution is applied to the forearm (between the wrist and elbow) or 
lower leg (between the knee and ankle) and the treated limb is exposed continuously to biting mosquitoes 
as the subject moves through mosquito-infested habitat. Biting mosquitoes are collected from treated and 
untreated skin (usually an exposed forearm) at regular intervals to determine mosquito biting rates and 
for species identification. This procedure is used to determine CPT, but percent repellency can also be 
calculated when a negative control is used. 


U.S. Environmental Protection Agency Test Guidelines 

The OPPTS guidelines have been developed for laboratory and field evaluation of pesticides and 
toxic substances and for acquiring test data submitted to the EPA for review under the Toxic 


2006 by Taylor & Francis Group, LLC 



108 


Insect Repellents: Principles, Methods, and Uses 


Substances Control Act (TSCA) (15 U.S.C. 2601) and the Federal Insecticide, Fungicide, and 
Rodenticide Act (FIFRA) (7 U.S.C. 136 et seq.). The product performance test guidelines contained 
in OPPTS 810.3700 describe specific methods for evaluating insect repellents and reflect the EPA’s 
minimum recommendations for developing reliable repellent product performance data. The draft 
guidelines are available electronically in portable document format (pdf) at http://www.epa.gov/ 
opptsfrs/publications/OPPTS_Harmonized/810_Product_Performance_Test_Guidelines/Drafts/810- 
3700.pdf 


Sources of Variation in Repellent Bioassays 
Abiotic Factors 

Many factors influence the outcome and interpretation of repellent bioassays. Skin-mediated effects 
comprise absorption and penetration of repellent on skin, but evaporation, abrasion (contact with 
clothing), washing or rinsing of treated surfaces, and perspiration also result in repellent loss. 15-18 These 
physical factors are difficult to control in a bioassay, but their contribution to experimental error can be 
minimized by random selection of test subjects, the use of appropriate sample sizes in bioassays, and by 
recognizing and avoiding pseudo replication. Loss of repellent by abrasion or by washing or rinsing from 
treated skin can be minimized by rigorous oversight of the test proceedings and by diligence on the part 
of the test subject. 

Light, temperature, humidity, and air quality at the testing venue are important environmental 
influences in repellents bioassays. ' ’ These factors can be manipulated to desired levels in the 
laboratory, but in nature their variation profoundly affects mosquito responses to repellent stimuli. 
Therefore, field bioassays should be standardized with respect to season, geographic location, and the 
time within the diel period in which observations are made. When this is not possible, tests should be 
designed so that estimates of important physical and climatic parameters are included as treatment 
variables in the statistical analysis. 

Additional environmental sources of variation in bioassays are repellent dose and exposure time 15 and 
test cage configuration. 7 ' 8 ’ 21 In the latter case, research suggests relationships between protection time, 
mosquito test population size, and the mosquito biting rate. However, investigations using different test 
cage configurations and mosquito population sizes 19 ' 21-24 have not led to a consensus regarding the 
optimal mosquito biting rate and density for repellency tests. One reason is that test cage shape and size 
and mosquito density effects vary between mosquito species. For Aedes aegypti, for example, repellent 
protection time is inversely related to cage size but is not affected by mosquito density; whereas, for 
Anopheles quadrimaculatus, protection time is short in large (125-L).cages with high mosquito densities 
(49 cm 3 per mosquito) and long in medium (65-L) cages with low mosquito densities (640 cm 3 per 
mosquito). 8 


Biotic Factors 

Biological factors in repellent bioassays consist of larval nutrition, carbohydrate availability to adult 
mosquitoes, age and parity in female mosquitoes, partial blood engorgement, and innate differences 
among repellent-treated test subjects. ’ ’ ’ An important behavioral factor that affects bioassay results 
is the timing and intensity of mosquito biting activity. 5 ' 27 Ignorance of temporal feeding patterns can 
compromise estimates of protection time for repellents that have extended activity, as can poor 
knowledge of biting rates. In screened cage tests, biting patterns can vary with the size of the cage, 
and this factor can affect the determination of repellency. 8 


© 2006 by Taylor & Francis Group, LLC 




Standard Methods for Testing Mosquito Repellents 


109 


Conclusions 

A comprehensive understanding of the parameters that affect repellent bioassays can minimize false 
positive responses in the early stages of repellent screening. Rigorous bioassay standards in the later 
stages of testing facilitate identification of the most promising new repellents and provide a sound basis 
for selecting new repellents for toxicology testing and evaluation in field tests. The selection of a 
repellents bioassay procedure should always be based on the biological relevance of the method and its 
capacity to yield precise experimental data. When these two outcomes are achieved, one can correlate the 
results from different bioassay techniques to obtain an accurate estimate of the repellency of 
any compound. 

References 

1. J. L. Robertson and H. K. Preisler, Pesticide Bioassays with Arthropods, Baca Raton: CRC Press, 
1992. 

2. World Health Organization [WHO], Report of the WHO informal consultation on the evaluation and 
testing of insecticides, World Health Organization, Control of Tropical Diseases, Pesticide Evaluation 
Scheme, Informal Consultation 96.1, Geneva, 1996. 

3. American Society for Testing and Materials [ASTM], Laboratory testing of non-commercial repellant 
formulations on the skin. ASTM-E951-94, 2000. 

4. American Society for Testing and Materials [ASTM], Field testing topical applications of compounds 
as repellents for medically important and pest arthropods. 1. Mosquitoes. ASTM-E393-94, 2000. 

5. H. K. Gouck and C. N. Smith, The effect of age and time of day on the avidity of Aedes aegypti, Fla. 
Entomol., 45, 93, 1962. 

6. C. N. Smith et ah. Factors affecting the protection period of mosquito repellents, USDA Tech. Bull., 
1258, 36, 1963. 

7. C. E. Schreck, Techniques for the evaluation of insect repellents: A critical review, Annu. Rev. 
Entomol, 22, 101, 1977. 

8. D. R. Barnard, Mediation of deet repellency in mosquitoes (Diptera: Culicidae) by species, age, and 
parity, J. Med. Entomol, 35, 340, 1998. 

9. J. A. Klun and M. Debboun, A new module for quantitative evaluation of repellent efficacy using 
human subjects, J. Med. Entomol, 37, 177, 2000. 

10. P. J. Weldon et ah, Benzoquinones from millipedes deter mosquitoes and elicit self-anointing in 
capuchin monkeys ( Cebus spp.), Naturwissenshaften, 90, 301, 2003. 

11. A. J. Klun et ah, A new in vitro bioassay system for discovery of novel human-use mosquito repellents, 
J. Am. Mosq. Control Assoc., 21, 64, 2005. 

12. M. D. Buescher et ah, The dose-persistence relationship of deet against Aedes aegypti, Mosq. News, 
43, 364, 1983. 

13. L. C. Rutledge et ah, Comparative sensitivity of representative mosquitoes (Diptera: Culicidae) to 
repellents, J. Med. Entomol, 20, 506, 1983. 

14. K. Posey and C. E. Schreck, An airflow apparatus for selecting female mosquitoes for use in repellent 
and attraction studies, Mosq. News, 41, 566, 1981. 

15. M. L. Gabel et ah. Evaporation rates and protection times of mosquito repellents, Mosq. News, 36, 141, 
1976. 

16. L. C. Rutledge et ah. Mathematical models of the effectiveness and persistence of mosquito repellents, 
J. Am. Mosq. Control Assoc., 1, 56, 1985. 

17. R. Gupta and L. C. Rutledge, Laboratory evaluation of controlled release repellent formulations on 
human volunteers under three climatic regimens, J. Am. Mosq. Control Assoc., 5, 52, 1989. 

18. L. M. Rueda et ah. Effect of skin abrasions on the efficacy of the repellent deet against Aedes aegypti, 
J. Am. Mosq. Control Assoc., 14, 178, 1998. 


2006 by Taylor & Francis Group, LLC 



110 


Insect Repellents: Principles , Methods, and Uses 


19. S. P. Frances et al.. Laboratory and field evaluation of deet CIC-4, and AI3-37220 against Anopheles 
dims (Diptera: Culicidae) in Thailand, J. Med. Entomol, 33, 511, 1996. 

20. W. G. Reinfenrath and T. S. Spencer, Evaporation and penetration from the skin, in Percutaneous 
Absorption: Mechanisms — Methods—Drug Delivery, R. L. Bronaugh and H. I. Maibach (Eds.), 
2nd ed., New York: Marcel Dekker, 1989, pp. 313-334. 

21. L. L. Lomax and P. Granett, Current laboratory procedures for the development of improved insect 
repellents at Rutgers-The State University, Proc. NJ Mosq. Exterm. Assoc., 58, 41, 1971. 

22. M. Bar-Zeev and D. Ben-Tamar, Evaluation of mosquito repellents, Mosq. News, 31, 56, 1971. 

23. A. A. Khan et al.. Insect repellents: Effect of mosquito and repellent related factors on protection time, 
J. Econ. Entomol., 68, 43, 1975. 

24. S. P. Frances et al„ Response of Anopheles dims and Aedes albopictus to repellents in the laboratory, 
J. Am. Mosq. Control Assoc., 9, 474, 1993. 

25. P. V. Wood, The effect of ambient humidity on the repellency of ethylhexanediol (‘6—12’) to Aedes 
aegypti Can. Entomol, 100, 1331, 1968. 

26. R. D. Xue and D. R. Barnard, Effects of partial blood engorgement and pretest carbohydrate 
availability on the repellency of deet to Aedes albopictus, J. Vector Ecol., 24, 111, 1999. 

27. R. D. Xue and D. R. Barnard, Human host avidity in Aedes albopictus: Influence of mosquito body 
size, age, parity, and time of day, J. Am. Mosq. Control Assoc., 12, 58, 1996. 


© 2006 by Taylor & Francis Group, LLC 



6 


Biometrics and Behavior in Mosquito 
Repellent Assays 


Donald R. Barnard and Rui-de Xue 

CONTENTS 

Introduction.Ill 

Physical and Biological Influences in Repellent Bioassays.112 

Mosquito Taxon.112 

Larval Rearing and Nutrition.112 

Adult Age, Oviparity, and Body Size.113 

Carbohydrate Availability.113 

Blood Feeding Patterns in Mosquitoes.113 

Mosquito Density, Landing Rate, and Repellency.113 

Attraction of Mosquitoes to Human Hosts.115 

Minimizing Variation in Repellent Bioassays.117 

Selection of Mosquito Taxa.117 

Selection of Human Test Subjects.117 

Selection of Mosquito Specimens.117 

Selection of Test Arena Configuration.118 

Management of the Repellent Bioassay Process.119 

Quantification of Repellency Responses.119 

Experimental Design.120 

Laboratory Bioassays.121 

Field Bioassays.121 

Conclusion.122 

References.122 


Introduction 

Humans use a variety of techniques for protection from arthropod bites. In the simplest case, one can 
avoid entering habitat that is infested with arthropod pests or disease vectors. Conversely, biting 
arthropods can be excluded from human living space by physical barriers, such as screens and nets, and 
by the use of building construction methods that prevent arthropod entry. 

Ill 


© 2006 by Taylor & Francis Group, LLC 




























112 


Insect Repellents: Principles , Methods, and Uses 


Tests for insect repellency are performed using a process called biological assay (bioassay for short). 1 
Repellent bioassays typically involve mosquitoes. In vitro repellent bioassays measure mosquito 
response to repellent on an inanimate surface, such as repellent-treated cloth, filter paper, and animal 
membrane, or to airborne repellents in an olfactometer. 2 ' 3 In vivo systems measure mosquito response to 
animal and human subjects that have been treated with a repellent. 4-8 

With both in vitro and in vivo bioassay systems, a stimulus is applied and a response to the stimulus by 
mosquitoes is observed. This process is repeated until an average response for the test population can be 
estimated with a desired level of precision. The stimulus can be a dose of repellent applied to human or 
animal skin or to an inanimate object. Typical responses comprise the number of mosquitoes that 
approach; land; land and probe; or land, probe, and bite the repellent-treated object. 

Procedural standards for in vivo evaluation of arthropod repellents in the laboratory and field have 
been published by the World Health Organization (WHO), 9 the American Society for Testing and 
Materials (ASTM), 10 ' 11 and the U.S. Environmental Protection Agency (EPA). 12 Two other method¬ 
ologies, the large screened-cage 8 ' 13 and the K&D module, 14 are commonly cited in the 
scientific literature. 


Physical and Biological Influences in Repellent Bioassays 

The results obtained in preliminary bioassays of a candidate repellent provide the basis for further study 
of the repellent, including toxicological evaluation and field testing. For this reason, it is important to 
minimize variation in repellency responses during the early stages of testing. Minimizing variation 
requires knowledge of the characteristics and limitations of the bioassay method that is used, as well as 
the capacity of the method to yield precise experimental data. 

Many physical and biological factors affect the outcome of a repellent bioassay. Some of this variation 
cannot be controlled and becomes part of the experimental error. Other sources of error can be identified 
and managed satisfactorily, particularly in the laboratory setting; known sources of variation can also be 
accounted for by blocking or the use of other experimental designs. 15 


Mosquito Taxon 

Species and genera of mosquitoes differ significantly in their responses to insect repellents. These 
differences appear to be independently inherited and unrelated to taxonomic distance. 16 Median effective 
dose (ED 5 o) values for deet vary by more than 300% among species in the same genus and by more than 
600% for species in different genera 16 ; similarly, intergeneric and intrageneric variation in the responses 
of different mosquito species to a wide range of repellents is not significantly different. 17 This means that 
repellency responses for one species of mosquito cannot be reliably inferred from those of another 
species, even among closely related taxa. 


Larval Rearing and Nutrition 

Overcrowding of mosquito larvae results in slow growth, small and/or irregular-sized adults, and low 
fecundity in females. “ Khan - observed similar protection times for deet against adult mosquitoes 
from the same larval population (cohort), whereas adults from different larval populations varied 
significantly in their responses to deet. In this regard, the most robust estimates of deet repellency will be 
obtained from bioassays that use adult mosquitoes from different larval cohorts. 


© 2006 by Taylor & Francis Group, LLC 



Biometrics and Behavior in Mosquito Repellent Assays 


113 


Adult Age, Oviparity, and Body Size 

In terms of repellent protection time, that provided by 25% deet against host-seeking female Aedes 
aegypti is not significantly affected by age or parity status, 13 nor do these variables interact to affect deet 
repellency. 22 Thus, one may expect consistent protection time responses to deet when considering the 
age and parity structure of the biting mosquito population. 

In contrast, mosquito age, oviparity, and body size (and the interactions of these factors) influence host 
attack rates ' ~ and, in some cases, repellency.” The interplay of host attack rates and repellent 
protection time 8 affects the risk of exposure to mosquito-borne disease agents. In Anopheles albimanus, 
the proportion of the population that is biting repellent-treated skin at the time of repellent failure is 
highest in 20-day-old parous females. 22 For nulliparous Aedes albopictus, host attack rates are higher in 
15- and 20-day-old (post-emergence) females than in 5- and 10-day-old females, regardless of body size, 
and in large compared with smaller females, regardless of age (Figure 6.1). 24 Deet (25%) repellency to 
large female Aedes albopictus is 2 h less than to small females. 24 

Carbohydrate Availability 

Sugar availability affects host-seeking behavior and blood feeding in Aedes aegypti . 25,26 Repellent 
protection times are 4.5 h against this species when females are provided sugar water ad libitum and 3.3 h 
when starved. 27 In Aedes albopictus, the pretest availability of 10% sucrose solution in screened cages 
increases host attack rates and the complete protection time (CPT) for 25% deet compared with females 
provided water only for 12 h before tests using the same repellent treatment (Table 6.1). 28 

Blood Feeding Patterns in Mosquitoes 

Repellent bioassays may require eight or more hours to complete. During this time, repeated observations 
for mosquito landing/biting activity are made at successive intervals within the diel (24 h) period. 
Mosquito host attack rates during such times can vary significantly, 24 ’ 29 ' 30 depending on the 
mosquito species. 

In afternoon tests against Aedes aegypti, CPT exceeds that in morning tests by 1,000%. 29 The 
differences are related to variations in body size, age, and oviparity and result in higher mean attack rates 
by large nulliparous females than by small nulliparous females. 31 Parous females 15 or more days old are 
more likely to attack human hosts than parous females less than 15 days old, whereas large-bodied, old, 
parous females exhibit the highest host attack rates overall. 

Five-day-old, large-bodied, nulliparous Aedes albopictus exhibit consistent host attack rate responses 
throughout the diel period. 24 This fact likely contributes to their widespread use in repellent 
bioassays. 817 Mosquitoes otherwise categorized according to parity or age can provide added rigor to 
tests of repellent effectiveness because of their high host attack rates. Fifteen- and twenty-day-old female 
Aedes albopictus tested between 1400 and 2000 h are one example. When calculated as a percentage of 
the mean daily host attack rate for 5-day-old females, attack rates for 15- and 20-day-old females range 
from — 18% to + 148%, depending on the time of day, and are lower (negative) only between 1000 and 
1200 h. This means that repellent bioassays commenced early in the day using 5-day-old female Aedes 
albopictus, and that last 6 h or longer, will overestimate CPT for 15- and 20-day-old females. 

Mosquito Density, Landing Rate, and Repellency 

Conventional test methods for mosquito repellents 9-11 assume a linear response to repellent dose and a 
constant level of mosquito biting activity. Despite these assumptions, repellency responses can be highly 
variable. 27 Differences between humans in their attractancy to mosquitoes 32 accounts for some of the 
variation, as do changed biting/landing rates caused by fluctuations in mosquito density and/or 


© 2006 by Taylor & Francis Group, LLC 



© 2006 by Taylor & Francis Group, LLC 


















Biometrics and Behavior in Mosquito Repellent Assays 


115 


TABLE 6.1 


Mean Landing Rates (in 60 s) of, and Repellent (25% deet) Protection Times 
Against, Nonblood-Fed Female Aedes albopictus Provided Water or Sucrose 
Solution (10% in Water) for 12 h Before Testing in Screened Cages 



Water 

Sucrose Solution 

Mosquito landing rate (+ SE) 

24.7 ( + 1.2) 

14.0 (±3.5) 

Repellent protection time (h + SE) 

6.2 ( + 0.3) 

8.2 (±0.3) 


endogenous cycles in the population. 24,29 An additional factor is repellent dose, the response to which by 
mosquitoes depends on taxon, testing arena, and mosquito density/biting pressure. In screened cage 
bioassays, Aedes aegypti responses to 25% deet are not significantly affected by mosquito landing rate; 
whereas, for Anopheles quadrimaculatus protection times are short when the rates are high and long 
when the rates are low. 8 

The protection time of deet against Aedes aegypti in screened cages varies with differences in 
mosquito density and repellent concentration. 8 The relationship between protection time and deet 
concentration (Table 6.2) shows that increases in repellent dose increase protection time, regardless of 
mosquito density. A high (45%) deet concentration provided longer repellency than expected, given the 
increase in repellency observed between 15% and 30% deet. In contrast, changes in protection time 
associated with increasing mosquito density are linear, irrespective of deet concentration (Table 6.2). 
Successive increases in mosquito density (up to and including 64 cm 3 cage volume per female mosquito) 
result in an approximately 50% reduction in protection time. 

In screened cages, there is a negative correlation between landing rates for Aedes aegypti and 
Anopheles quadrimaculatus on a human subject and repellent (25% deet) protection time. 8 Regression 
analysis indicates that a significant portion of the repellency responses for both species can be explained 
on the basis of mosquito landing rate 8 and that estimated protection times (Figure 6.2) range from 4.6 to 
6.2 h, when Aedes aegypti landing rates are 62 to 6 per half minute, respectively, and from 1.8 to 6.5 h 
when Anopheles quadrimaculatus landing rates are from 55 to 2 per half minute, respectively. 

Attraction of Mosquitoes to Human Hosts 

Mosquitoes use vision, heat, and host emanations to locate their prey. 33-35 Fluman emanations that attract 
hungry mosquitoes include carbon dioxide, “ carboxylic acids, and lactic acid/ Mixtures of 


TABLE 6.2 


Mean (+ SE) Protection Time from Bites of Aedes aegypti Using Three Concentrations 
of Repellent (Deet) and Three Densities of Mosquitoes in Screened Cages 


Deet 

Concentration (%) 

Mean (+ SE) Repellent Protection Time (min) 

Mosquito Density 3 

Low 

Medium 

High 

15 

290 (±40) 

230 (±40) 

130 (±20) 

30 

320 (±10) 

260 (±26) 

160 (±10) 

45 

530 (±20) 

360 (±12) 

290 (±52) 


a Number of female mosquitoes per test cage (unit of test cage volume per female). Low: 200 
(640 cm 3 per female); medium: 1,000 (128 cm 3 per female); high: 2,600 (49 cm 3 per female). 


© 2006 by Taylor & Francis Group, LLC 









116 


Insect Repellents: Principles , Methods, and Uses 




Mosquito landings in 30 seconds 


FIGURE 6.2 Observed and estimated repellent protection times against Aedes aegypti and Anopheles quadrimaculatus in 
relation to mosquito landing rates in screened cages. Broken lines are the upper and lower 95% confidence limits for 
estimated repellent protection time. 


emanations, such as carbon dioxide and L-lactic acid, are highly attractive to Aedes albopictus , as are 
combinations of these substances with a variety of sulfides, ketones, and halogenated compounds. 43 

Mosquito attraction responses can vary widely among the human subjects participating in a 
repellents bioassay. They also depend on the species of mosquito that is being used. 32 Human hosts, 
for example, are highly or moderately attractive to Anopheles quadrimaculatus , Anopheles 
freeborni, and Culex salinarius, but less attractive to Anopheles crucians and Culex nigripalpus. 
For a given human subject, attraction responses vary depending on the body region to which hungry 
mosquitoes are exposed. 32 Early workers sought to explain these differences on the basis of skin 
temperature, gender, age, and other simple effects. 4445 We know now that host-finding in 
mosquitoes involves a complex of host-mediated and mosquito-based behavioral events and physio¬ 
logical processes 46 


© 2006 by Taylor & Francis Group, LLC 





Biometrics and Behavior in Mosquito Repellent Assays 


117 


Minimizing Variation in Repellent Bioassays 
Selection of Mosquito Taxa 

As noted earlier, species and genera of mosquitoes differ in their responses to repellents and human test 
subjects in a manner unrelated to taxonomic distance. 16 ' 32 The accurate determination of repellency thus 
requires that candidate compounds be tested against the target species. 16 In this regard, practical 
recommendations for the use of mosquito repellents should be based on laboratory and field bioassays 
that use species of known pest or vector importance. 47 This can be accomplished by collecting, rearing, 
and testing field specimens in the laboratory, or by testing wild populations in the field, although outdoor 
testing of repellents in areas with endemic mosquito-borne pathogens is accompanied by the risk of 
human infection. 

In general, the probability of detecting repellency is increased by using repellent-sensitive species, 
such as Aedes taeniorhynchus or Culex pipiens.' 1 In contrast, the identification of broad-spectrum 
repellency requires the use of mosquito species, such as Aedes aegypti and Anopheles albimanus , that 
have a low sensitivity to repellents. 17 


Selection of Human Test Subjects 

Gilbert et al. 44 showed that female subjects, on average, were less attractive to Aedes aegypti than male 
subjects. These workers observed an inverse (but not significant) correlation between attractiveness to 
mosquitoes and protection time in male subjects, when using 5% deet, but not in female subjects. Neither 
body weight, age, nor skin color/temperature affected repellent protection time, regardless of gender, 
although low skin moisture production in females was significantly related to increased repellent 
protection time. The results of Gilbert et al. 44 generally agreed with those of earlier studies, 48 ’ 30 but later 
analysis of their data by Rutledge 49 suggested that differences in deet protection times for men and 
women were not proved. 

One concern in field bioassays of repellents is the variance of estimates of mean mosquito biting rate. 
Typically, the innate attractiveness of human subjects to mosquitoes ranges from 30% to 70%, 32 thus, 
estimates of the biting rate can be imprecise, particularly when based on small sample size. Increasing the 
numbers of test subjects improves precision but the resources required to do so quickly exceed practical 
limits. As an alternative to large sample sizes, Barnard et al. 50 suggested that test subjects be selected 
according to their comparative attractiveness to mosquitoes. This factor can be determined with an 
olfactometer, 51-53 or by other means. Subjects selected for use in repellents bioassays would be those 
individuals with an attractiveness index within 1 or 2 standard deviations of the mean mosquito 
attractiveness index for the test population. 


Selection of Mosquito Specimens 

Mosquitoes selected for testing in laboratory repellent bioassays should be of equal age, sex, size, and 
vigor, and should be otherwise manipulated as little as possible prior to testing. Techniques for this 
purpose that involve the use of carbon dioxide, low temperature, anesthesia, or aspiration 54-56 subject 
mosquitoes to chemical exposure, drying, and temperature extremes and can induce morbidity-related 
behavioral changes in the test population with resultant variation in bioassay results. 

One method for minimizing trauma to the mosquitoes used for a bioassay is to attract and capture host¬ 
seeking females. This can be accomplished with the apparatus described by Posey and Schreck 57 that 
encloses a stock cage of mosquitoes and combines air flow and human odor to attract hungry females into 
a transfer module. Adjusting the flow rate of air through the apparatus allows one to count mosquitoes as 


2006 by Taylor & Francis Group, LLC 



118 


Insect Repellents: Principles, Methods, and Uses 


they exit the stock cage and enter the transfer module. The module is then placed inside a test cage, 
opened, and the mosquitoes released. Because this apparatus extracts only host-seeking female 
mosquitoes from the stock cage mosquito population, one can pre-select the size of the biting mosquito 
population before a test is made. This process can be repeated to produce equivalent mosquito biting 
pressures in subsequent bioassays. 

Selection of Test Arena Configuration 

One advantage of the dose-response (small cage) testing method is that it provides a measure of 
repellency at the level of the mosquito population median and/or other percentile(s) of interest. These 
measures are essentially independent of the size of the mosquito populations tested. 3 However, skill is 
required in the design of experiments that use small cage testing methods as treatment effects can be 
confounded with “edge effects,” 58 the latter as a consequence of position (of a feeding port or module). In 
addition, multiple replicates of treatments on the same human subject do not provide a basis for 
comparison of treatments among different subjects, 59 the attractiveness or repellency of which, to 
mosquitoes, can be highly variable. 32 

The determination of protection times using the screened cage method is based on the responses 
of mosquitoes in the upper extreme of the frequency distribution for repellent tolerance rather than 
on the mean response of the population. This technique does not measure the ED 50 of the test 
repellent by the mosquito population or other percentiles of interest. 16 Additionally, it confounds 
variation in repellent activity (per unit concentration applied) with the rate of repellent loss from the 
skin. 60 ’ 61 

In screened cage tests, test cage configuration affects repellency responses. 8 ' 13 However, studies of test 
cage configuration ’ ’ ’ ’ have not led to a consensus regarding an optimal configuration. This is 
because the effects of test cage shape and size vary depending on the species of mosquito under study. 
For Aedes aegypti, CPT is inversely related to cage size, whereas for Anopheles quadrimaculatus, 
protection time is shortest in 125-L cages with 49 cm 3 of cage volume per female and longest in 65-L 
cages with 640 cm 3 of cage volume per mosquito. 8 

Barnard et al. 8 used regression analyses to identify combinations of cage size and mosquito density 
that posed a range of challenges (from least to most rigorous) to the repellency of 25% deet in a 
laboratory test with Aedes aegypti and Anopheles quadrimaculatus. For both species, estimated CPTs 
were proportional to mosquito density, but showed a curvilinear relationship to cage size. 
Accordingly, the shortest protection times against Aedes aegypti were observed in large cages with 
high mosquito densities, where longer CPTs were associated with small cages and low mosquito 
densities (640 cm 3 of cage volume per female) (Figure 6.3). For Anopheles quadrimaculatus, large 
cages with high mosquito densities resulted in short CPTs; medium cages with low mosquito densities 
resulted in long CPTs. 

An important consideration when accepting or rejecting cage size and mosquito density parameters 
is temporal variation in the host avidity pattern. 24 In an attempt to characterize this phenomenon, 
Barnard et al. 8 calculated deviations in the mosquito biting rate in three different cage sizes at 0800, 
1200, and 1600 h as a percentage of the mean biting rate, and used the deviations to select or reject 
cage size and mosquito density conditions (Table 6.3). Based on a +25% deviation from the mean 
biting rate, large cages with low mosquito densities, medium cages with medium mosquito densities, 
and small cages with high mosquito densities would not be used in repellent tests with Aedes aegypti 
or Anopheles quadrimaculatus. The +25% deviation also excluded the use of medium cages with 
low densities of Aedes aegypti or small cages with low densities of Anopheles quadrimaculatus. A 
deviation of ± 10% indicates that large cages with high mosquito densities and small cages with 
medium mosquito densities would be acceptable for use in repellent bioassays with Aedes aegypti and 
Anopheles quadrimaculatus, as would medium cages with high mosquito densities in assays with 
Anopheles quadrimaculatus. 


© 2006 by Taylor & Francis Group, LLC 


Biometrics and Behavior in Mosquito Repellent Assays 


119 


Female density 



Large Medium Small 

Cage size 


FIGURE 6.3 Estimated repellent (25% deet) protection time in relation to cage size (large, medium, small = 125, 65, and 
27 L, respectively) and the density (high, medium, and low=49, 128, and 640 cm 3 of cage volume per female mosquito) of 
female mosquitoes in screened cages. 


Management of the Repellent Bioassay Process 
Quantification of Repellency Responses 

Repellency responses in mosquitoes can be quantified in terms of the effective dose (ED), CPT, and/or 
percent repellency (%R). The ED method is based on the dose-response data obtained according to 
ASTM E951-94 10 and is used to calculate the median (ED 5 o) and other ED percentiles of interest for a 
repellent. 17 ' 64 The CPT from mosquito bite is that time elapsed between application of a repellent on the 


2006 by Taylor & Francis Group, LLC 


















120 


Insect Repellents: Principles , Methods, and Uses 


TABLE 6.3 


Deviations from Mean Landing Rate on Human Subjects by Aedes aegypti ( Aa ) and Anopheles 
Quadrimaculatus (Aq) at Three Times of Day, in Three Different Sized Cages, Using Three 
Densities of Female Mosquitoes 


Mosquito 
Density and 


Cage Size (volume) 



Large (125 L) 

Medium (65 L) 

Small (27 L) 


Time (h) 

Aa Aq 

Aa 

Aq 

Aa 

Aq 

Low ( 640 cm 3 
0800 

cage volume per female) 

-27 -37 

n 

6 

6 

-48 

1200 

5 34 

-57 

-13 

-12 

4 

1600 

22 3 

46 

7 

6 

44 

Medium (128 
0800 

cm 3 cage volume per female) 

-4 -5 

-19 

25 

-7 

0 

1200 

-15 -4 

-32 

4 

8 

8 

1600 

19 9 

51 

-29 

-1 

-8 

High (49 cm 3 
0800 

cage volume per female) 

7 -4 

-11 

7 

-5 

28 

1200 

-9 -2 

-13 

-6 

-25 

1 

1600 

2 6 

24 

-1 

30 

-29 


skin and the first mosquito bite on the treated skin, or the time between repellent application and the 
observation time immediately preceding the first bite. Percent repellency is a quotient, comprising the 
difference (at the same point in time) between mosquito biting rates on untreated and repellent treated 
skin, divided by the biting rate on untreated skin, multiplied by 100. A negative control is required for 
calculation of %R. 

The ED method is used to characterize repellency in insectary-reared and wild mosquitoes; however, 
the data for this purpose are acquired in the laboratory. Complete protection time and %R can be used to 
describe repellency responses in laboratory and field bioassays. Evaluations based on ED measure the 
inherent repellency of a chemical with no consideration of how long the chemical will produce 
repellency. CPT and some uses of %R measure what is essentially a combined statistic for inherent 
repellency and duration. Therefore, a hypothetically good repellent might have a higher ED, but be a 
powerful active ingredient because its volatility and skin absorption are low. On the other hand, a volatile 
repellent with a low ED might be a poor product because it disappears from the skin too quickly. 


Experimental Design 

Pre-test conditions that favor an objective outcome in repellent bioassays include the randomization of 
test subjects and of treatments among test subjects, adequate replication, and the use of negative 
(untreated) and/or positive (treated) controls. A negative control can be the biting rate observed on the 
untreated forearm, leg, or other body part of a subject when exposed to a population of mosquitoes. In 
laboratory tests, a positive (treated) control can be the biting rate on one forearm of a subject that has 
been treated with a known repellent (25% deet in ethanol), compared with the biting rate on the same 
subject’s remaining forearm that has been treated with a repellent of unknown efficacy, and each arm 
exposed separately to a population of mosquitoes. Positive controls are used to determine the 
comparative efficacy of two repellents on the same test subject but for the reasons described below 
should not be used in field bioassays. If positive and negative controls are used in the same bioassay, they 
should be allocated to separate test subjects. 

The effects of skin-mediated sources of variation (absorption, penetration, evaporation, perspiration) 
in bioassays can be minimized by random selection of test subjects, the use of an adequately-sized test 
population, and by proper replication of treatments. Loss of repellent by abrasion or by washing/rinsing 


© 2006 by Taylor & Francis Group, LLC 



Biometrics and Behavior in Mosquito Repellent Assays 


121 


from treated skin can be minimized by careful oversight and management of the test proceedings and by 
diligence on the part of the test subject. 

Rutledge 49 noted three shortcomings of the methods described in ASTM 939-94 11 for data obtained in 
field repellent bioassays using the incomplete block design (IBD). The first concerned the design itself, 
correction of which involves analysis of data from a “resolvable balanced IBD,” rather than the balanced 
IBD. The second shortcoming concerned inefficient evaluation of inter-block information, and the third, 
improper use of adjusted means for estimating treatment means. Rutledge 49 provides details for 
correcting each problem. 

Laboratory Bioassays 

When determining repellency in the laboratory, at least two techniques can be used to address time- 
of-observation-based systematic errors. 24 The first involves the use of 5-day-old nulliparous females in 
bioassays and the application of test results and inferences to only this group of mosquitoes. The second 
technique is based on the assumption of equivalent host avidity throughout a bioassay and requires 
commencement and completion of the test between 1400 and 2000 h (when sunrise is at 0600 h). When 
this is not possible (for example, when bioassay times exceed 6 h), a single test can be divided into two 
phases (early and late) of 3-6 h duration each, with the order of the phases in each test selected 
at random. 

Field Bioassays 

ASTM E939-94 11 prescribes the side-by-side comparison of repellents on the same test subject. Results 
obtained in this manner can be misleading because the presence of repellent on one arm of a subject 
affects the mosquito landing rate (and apparent repellency) on the opposite arm of the same subject. 50 
When the landing rate on negative controls is used as the reference point, this effect is manifested as a 
lower landing rate on repellent treated subjects early in tests and a higher landing rate on the same 
subjects late in tests. An example is shown for para-menthane-3,8-diol (PMD) (against Aedes 
taeniorhynchus in the Everglades National Park [U.S.A.]), the landing rate for which, on repellent 
treated subjects, differs from the landing rate on control subjects in periods 3 and 7 (approximately 3 and 
7 h after repellent application) by —8% and +22%, respectively (Table 6.4). The difference equates to 
an actual %R of 95.6% and 60.0%, respectively, compared with calculated %Rs of 85% and 79%, and 
leads to the underestimation of PMD repellency (by 11%) in period 3 and overestimation of repellency 
(by 19%) in period 7. Such interactions result in confounding of treatment (repellent) effects, biased 
estimates of mosquito biting rate, and faulty estimation of repellency responses. One solution to this 
problem is to use one repellent per test subject and one or more negative controls in each bioassay. 9 


TABLE 6.4 


Mean Landing Rates for Aedes Taeniorhynchus on the Untreated Forearms of Human Subjects Whose 
Opposite Forearms Had Been Treated with Deet. KBR3023, IR3535, or />ara-menthane-3,8-diol (PMD), 
Calculated as a Percent of the Mean Landing Rate on the Forearms of (Ethanol only) Control Subjects 


Approximate Number of Hours 


Percent of Control 


Following Repellent Application 

Deet 

KBR3023 

IR3535 

PMD 

1 

63 

72 

76 

60 

3 

85 

82 

78 

92 

5 

83 

91 

102 

87 

7 

84 

103 

98 

122 


2006 by Taylor & Francis Group, LLC 



122 


Insect Repellents: Principles, Methods, and Uses 


Field tests should be made comparable with respect to season, geographic location, and the time of the 
diel period in which observations are made. When this is not possible, estimates of the important physical 
and climatic parameters need to be included as treatment variables in the experimental design. In this 
same regard, when mosquito biting rate and the assessment of repellency is based on the capture of 
mosquitoes attacking human volunteers, bioassays should be timed to exploit the biting cycle of the 
target mosquito species. Test subjects should be spaced at least 10 m apart and rotated in location in a 
randomized manner throughout the experiment to minimize positional bias. 9 


Conclusion 

Unfortunately, no single repellent bioassay system provides a definitive measure of repellent 
effectiveness against mosquitoes. In fact, there is probably no “best” repellent bioassay system. 
Nevertheless, it is important to know the suitability of a given system to the natural history and behavior 
of the taxon under study, as well as the biological meaning of the species’ response to repellent within the 
physical context of the bioassay system. Given these conditions, and the minimization of external sources 
of variation, the repellent bioassay system should provide precise and repeatable measurements of 
repellency. A reliable judgment of repellent effectiveness can be made on this basis and verified by 
comparison with results from other repellent bioassay systems. 

References 

1. J. L. Robertson and H. K. Preisler, Pesticide Bioassays with Arthropods, Boca Raton, FL: CRC Press, 
1992. 

2. H. S. Lai, S. Ginocchio, and E. J. Hawrylewicz, Procedure for bioassaying mosquito repellents in 
laboratory animals, Proc. Soc. Exper. Biol. Med., 113, 770, 1963. 

3. L. C. Rutledge, M. A. Moussa, and C. J. Belletti, An in vitro blood-feeding system for quantitative 
testing of mosquito repellents, Mosq. News, 36, 283, 1976. 

4. S. P. Frances et al.. Response of Anopheles dirus and Aedes albopictus to repellents in the laboratory, 
J. Am. Mosq. Control Assoc., 9, 474, 1993. 

5. J. T. Chou, P. A. Rossignal, and J. W. Ayres, Evaluation of commercial insect repellents on human 
skin against Aedes aegypti (Diptera: Culicidae), J. Med. Entomol., 34, 624, 1997. 

6. J. A. Hill et al., Evaluation of mosquito repellents on the hairless dog, Mosq. News, 39, 307, 1979. 

7. L. C. Rutledge et al.. Evaluation of the laboratory mouse model for screening topical mosquito 
repellents, J. Am. Mosq. Control Assoc., 10, 565, 1994. 

8. D. R. Barnard et al.. Mosquito density, biting rates and cage size effects on repellents tests, Med. Vet. 
Entomol, 12, 39, 1998. 

9. World Health Organization (WHO), Report of the WHO informal consultation on the evaluation and 
testing of insecticides, World Health Organization, Control of Tropical Diseases, Pesticide Evaluation 
Scheme. Informal Consultation 96.1, Geneva, 1996. 

10. American Society for Testing and Materials (ASTM), Laboratory testing of non-commercial repellant 
formulations on the skin, ASTM-E951-94, 2000. 

11. American Society for Testing and Materials (ASTM), Field testing topical applications of 
compounds as repellents for medically important and pest arthropods. 1. Mosquitoes, ASTM-E393- 
94, 2000. 

12. United States Environmental Protection Agency (USEPA), Office of Prevention, Pesticides, and Toxic 
Substances, Product performance test guidelines, Insect repellents for human skin and 
outdoor premises, 810.3700, 1999, http://www.epa.gov/opptsfrs/publications/OPPTS_Harmonized/ 
810_Product_Performance_Test_Guidelines/Drafts/810-3700.pdf (accessed 2006). 

13. C. E. Schreck, Techniques for the evaluation of insect repellents: a critical review, Ann. Rev. Entomol., 
22, 101, 1977. 


© 2006 by Taylor & Francis Group, LLC 




Biometrics and Behavior in Mosquito Repellent Assays 


123 


14. J. A. Klun and M. Debboun, A new module for quantitative evaluation of repellent efficacy using 
human subjects, J. Med. Entomol., 37, 177, 2000. 

15. W. G. Cochran and G. M. Cox, Experimental Designs , 2nd ed.. New York: Wiley, 1957. 

16. L. C. Rutledge et al., Comparative sensitivity of mosquito species and strains to the repellent diethyl 
toluamide, J. Med. Entomol., 14, 536, 1978. 

17. L. C. Rutledge et al.. Comparative sensitivity of representative mosquitoes (Diptera: Culicidae) to 
repellents, J. Med. Entomol., 20, 506, 1983. 

18. T. Ikshoji. The influence of larval breeding conditions on fecundity of Culex pipiens fatigans Weid, 
WHO Vector Contrib., 135, 65. 1965. 

19. A. A. Khan et al., Increased attractiveness of man to mosquitoes with increased eccrine sweating, 
Nature, 223, 859, 1969. 

20. M. J. Klowden. Factors influencing multiple host contacts by mosquitoes during a single gonotrophic 
cycle. Misc. Public. Entomol. Soc. Am., 68, 29, 1988. 

21. A. A. Khan, Gross variation in the response to man among yellow fever populations in the laboratory, 
J. Econ. Entomol, 62, 96, 1969. 

22. D. R. Barnard, Mediation of deet repellency in mosquitoes (Diptera: Culicidae) by species, age, and 
parity, J. Med. Entomol, 35, 340, 1988. 

23. M. M. Klowden, J. L. Blackmer, and G. M. Chambers, Effects of larval nutrition on the host-seeking 
behavior of adult Aedes aegypti mosquitoes, J. Am. Mosq. Control Assoc., 4, 73, 1988. 

24. R. D. Xue and D. R. Barnard, Human host avidity in Aedes albopictus: influence of mosquito body 
size, age, parity, and time of day, J. Am. Mosq. Control Assoc., 12, 58, 1996. 

25. W. A. Foster and F. Eischen, Frequency of blood-feeding in relation to sugar availability in Aedes 
aegypti and Anopheles quadrimaculatus (Diptera: Culicidae), Ann. Entomol. Soc. Am., 80, 103, 1987. 

26. M. M. Klowden et al.. Effects of carbohydrate ingestion on the pre-oviposition behavior of the 
mosquito Aedes aegypti (L.), Bull Soc. Vector Ecol., 15, 59, 1990. 

27. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Insect repellents: effect of mosquito and repellent 
related factors on protection time, J. Econ. Entomol., 68, 43, 1975. 

28. R. D. Xue and D. R. Barnard, Effects of partial blood engorgement and pretest carbohydrate 
availability on the repellency of deet to Aedes albopictus, J. Vector Ecol., 24, 111, 1999. 

29. H. K. Gouck and C. N. Smith, The effect of age and time of day on the avidity of Aedes aegypti, Fla. 
Entomol., 45, 93, 1962. 

30. C. N. Smith et al.. Factors affecting the protection period of mosquito repellents, USDA Tech. Bull 
1258. 

31. R. Xue, D. R. Barnard, and C. E. Schreck, Effect of body size. Parity, and age of Aedes albopictus on 
human host attack rates and the repellency of deet, J. Am. Mosq. Control. Assoc., 11, 50, 1995. 

32. C. E. Schreck, D. L. Kline, and D. A. Carlson, Mosquito attraction to substances from the skin of 
different humans, J. Am. Mosq. Control Assoc., 6, 406, 1990. 

33. G. D. Price, N. Smith, and D. A. Carlson, The attraction of female mosquitoes ( Anopheles 
quadrimaculatus Say) to stored human emanations in conjunction with adjusted levels of relative 
humidity, temperature, and carbon dioxide, J. Chem. Ecol., 5, 383, 1979. 

34. W. Takken, The role of olfaction in host-seeking of mosquitoes: a review. Ins. Sci. Applic., 12,287,1991. 

35. W. Takken and B. G. J. Knols, Odor-mediated behavior of Afrotropical malaria mosquitoes, Ann. Rev. 
Entomol., 44, 131, 1999. 

36. C. E. Schreck, H. K. Gouck, and C. N. Smith, An improved olfactometer for use in studying mosquito 
attractants and repellents, J. Econ. Entomol, 60, 1188, 1967. 

37. M. T. Gillies, The role of carbon dioxide in host-finding by mosquitoes (Diptera: Culicidae) a review, 
Bull. Entomol Res., 70, 525, 1980. 

38. M. Geier, O. J. Bosch, and J. Boeckh, Ammonia as an attractive component of host odour for the yellow 
fever mosquito Aedes aegypti, Chem. Sens., 24, 647, 1999. 

39. F. Acree et al., L-Lactic acid: a mosquito attractant isolated from humans, Science, 161, 1346, 1968. 

40. O. J. Bosch, M. Geier, and J. Boeckh, Contribution of fatty acids to olfactory host-finding of female 
Aedes aegypti, Chem. Sens., 25, 323, 2000. 


2006 by Taylor & Francis Group, LLC 



124 


Insect Repellents: Principles, Methods, and Uses 


41. T. Dekker et al., L-Lactic acid: a human-signifying host cue for the anthropophilic mosquito Anopheles 
gambiae, Med. Vet. Entomol., 16, 91, 2002. 

42. U. R. Bernier et al., Analysis of human skin emanations by gas chromatography/mass spectrometry, 2. 
Identification of volatile compounds that are candidate attractants for the yellow fever mosquito. Anal. 
Chem., 72, 7474, 2000. 

43. U. R. Bernier et al., Chemical composition that attract arthropods, U.S. Patent 6,800,279, 2004. 

44. I. H. Gilbert, H. K. Gouck, and N. Smith, Attractiveness of men and women to Aedes aegypti and relative 
protection time obtained with deet, Fla. Entomol., 49, 53, 1966. 

45. A. A. Khan, Mosquito attractants and repellents, in Chemical Control of Insect Behavior, H. H. Shorey 
and J. J. McKelvey (Eds.), New York: Wiley, pp. 155-176, 1977. 

46. A. N. Clements, Sensory Reception and Behavior, The Biology of Mosquitoes, Vol. 2, Wallingford: 
CABI Publishing, 1999. 

47. D. R. Barnard and R. D. Xue, Laboratory evaluation of mosquito repellents against Aedes albopictus 
Culex nigripalpus, and Ochlerotatus triseriatus, J. Med. Entomol., 41, 726, 2004. 

48. H. K. Gouck and M. C. Bowman, Effect of repellents on the evolution of carbon dioxide and moisture 
from human arms, J. Econ. Entomol, 52, 1157, 1959. 

49. L. C. Rutledge, Some corrections to the record on insect repellents and attractants, J. Am. Mosq. Control 
Assoc., 4, 414, 1988. 

50. D. R. Barnard et al., Repellency of IR3535, KBR3023, para-menthane-3,8-diol, and deet to Black Marsh 
mosquitoes (Diptera: Culicidae) in the Everglades National Park, J. Med. Entomol, 39, 895, 2002. 

51. A. A. Khan, Effects of repellents on mosquito behavior, Quaest. Entomol, 1, 1, 1965. 

52. K. H. Posey, D. R. Barnard, and C. E. Schreck, Triple cage olfactometer for evaluating mosquito 
(Diptera: Culicidae) attraction responses, J. Med. Entomol, 35, 330, 1998. 

53. D. J. Mauer and W. A. Rowley, Attraction of Culex pipiens (Diptera: Culicidae) to flower volatiles, 
J. Med. Entomol, 36, 503, 1999. 

54. R. I. Harris, R. A. Hoffman, and E. D. Frazer, Chilling vs. other methods of immobilizing flies, J. Econ. 
Entomol, 58, 379, 1965. 

55. C. M. Gjullin and V. D. Bevill, Insect chilling table, J. Med. Entomol, 9, 266, 1972. 

56. L. A. Magnerelli, A portable battery powered aspirator for mosquito collection, J. Med. Entomol, 12, 
308, 1975. 

57. K. H. Posey and C. E. Schreck, An airflow apparatus for selecting female mosquitoes for use in repellent 
and attraction studies, Mosq. News, 41, 566, 1981. 

58. T. R. E. Southwood and P. A. Henderson, Ecological Methods, 3rd Ed., Oxford: Blackwell Science, 

2000 . 

59. R. Mead, The Design of Experiments, Statistical Principles for Practical Application, Cambridge: 
Cambridge University Press, 1988. 

60. C. F. Curtis et al., The relative efficacy of repellents against mosquito vectors of disease, Med. Vet. 
Entomol., 1, 109, 1987. 

61. L. C. Rutledge et al., Mathematical models of the effectiveness and persistence of mosquito repellents, 
J. Am. Mosq. Control Assoc., 1, 56, 1985. 

62. M. Bar-Zeev and D. Ben-Tamar, Evaluation of mosquito repellents, Mosq. News, 31, 61, 1971. 

63. L. L. Lomax and P. Granett, Current laboratory procedures for the development of improved insect 
repellents at Rutgers—The State University, Proc. NJMosq. Exterm. Assoc., 58, 41, 1971. 

64. M. D. Buescher et al.. The dose-persistence relationship of deet against Aedes aegypti, Mosq. News, 43, 
364,1983. 


© 2006 by Taylor & Francis Group, LLC 



7 


Animal Models for Research and Development of 
Insect Repellents for Human Use 


Louis C. Rutledge and Raj K. Gupta 


CONTENTS 

Introduction.126 

History and Development.127 

Arthropod Target Species.127 

Vertebrate Species.128 

Interspecific Differences.129 

Skin Temperature.129 

Skin Permeability.129 

Blood Content and Circulation.129 

Eccrine and Apocrine Sweat Glands.130 

Sebaceous Glands.130 

Hair.130 

Technique.131 

Humane Treatment of Experimental Animals.131 

Preparation and Treatment.131 

Test Methods.132 

Test Population Size.132 

Observing and Recording Test Data.133 

Methods of Observing and Recording.133 

Recording Protection Time.133 

Experimental Design and Data Analysis.134 

Null Treatment (Control Experiment).134 

Protection Time Models.134 

Protection Time as a Random Variable from a Normal Distribution.134 

Observations from Truncated and Censored Distributions.135 

Failure Time Data.135 

Singularities in Catastrophe Data.135 

Analysis of Variance.136 

Paired Observations.136 

One-Way and Multi-Way Experimental Designs.136 

Balanced Incomplete Block Designs.136 


125 


© 2006 by Taylor & Francis Group, LLC 



































126 


Insect Repellents: Principles, Methods, and Uses 


Bioassay Methods.136 

No-Choice and Free-Choice Designs.137 

Effective Dose.138 

Persistence.138 

Extrapolation to Humans.139 

Material Standards and Comparative Observations.139 

Statistical Adjustment of Data.140 

Correction Terms.140 

Correction Factors.141 

Curve Fitting.141 

Conclusion.141 

References.142 


Introduction 

Experimental animals have been widely used in basic and applied research leading to the commercial 
production and sale of insect repellents (Figure 7.1). The present review is concerned with the use of 
animal subjects in research and development of repellents intended for human use, but studies using 
human subjects and studies of repellents intended for veterinary use are cited where relevant. 
Toxicological studies have been excluded from consideration because toxicological tests of repellents 



FIGURE 7.1 Laboratory test of repellents on infant mice in progress. Five treated mice were confined in hardware cloth 
boxes inside a test cage containing 100 female mosquitoes, and biting activity was recorded at 2-rnin intervals for 20 min. 
(From Letterman Army Institute of Research, San Francisco, CA.) 


© 2006 by Taylor & Francis Group, LLC 



















Animal Models for Research and Development of Insect Repellents for Human Use 


127 


on animals do not involve the target insects. Studies in which tissues or organs such as blood or skin are 
used in lieu of the whole animal and studies of repellents intended for use against nonarthropod forms 
such as schistosome cercariae or leeches are also excluded. 

The preparation of the review included the identification and assembly of 53 scientific papers published 
over the period 1926-2004 reporting studies utilizing experimental animals in research and development 
of repellents. In addition, the review used 11 scientific papers published by the authors and colleagues of 
the former Letterman Army Institute of Research, Presidio of San Francisco, California, over the period 
1980-1997 in an unofficial program of research on the use of laboratory animals in repellent testing. 


History and Development 

Perhaps the first recorded use of animals to test repellents intended for human use was that of Kawamura 
in 1926. Kawamura used monkeys, guinea pigs, and rabbits in tests of repellents against Leptotrombi- 
dium akamushi (Brumpt) (Acari: Trombiculidae) in Japan. 1 In this survey, no subsequent example was 
found until that of Lindquist et al. in 1944, who used chickens in tests of repellents against 
Ctenocephalides fells (Bouche), Ctenocephalides canis (Curtis), and Echidnophaga gallinacea 
(Westwood) (Siphonaptera: Pulicidae). 2 

The number of studies using animals in research and development of repellents for human use 
increased slowly through the 1940s (5 studies found), 1950s (7 studies found), 1960s (9 studies found), 
1970s (9 studies found), and 1980s (12 studies found), decreasing thereafter in the 1990s (7 studies 
found), and 2000s (6 studies found). The figure given for the 2000s is a projection, being prorated from 
the five years 2000-2004 to the whole decade. To avoid bias, 11 papers published by the authors and 
colleagues of the Letterman Army Institute of Research from 1980 to 1997 were excluded. The rise of the 
modern animal rights movement from the 1970s to the present 3 is a possible contributing factor to the 
decline in the number of studies conducted in the 1990s and 2000s. It is not possible now to know either if 
the decline will continue or for how long it will decline. In any case, a variety of repellent test systems 
utilizing a variety of animal species in tests of a variety of repellents against a variety of arthropods had 
been described in the scientific literature by 2004. The sections that follow present the principles and 
procedures demonstrated in this body of literature as currently understood. 


Arthropod Target Species 

Species belonging to eight arthropod families were targeted in the 53 studies reviewed: Argasidae (soft 
ticks), Ixodidae (hard ticks), Trombiculidae (chigger mites), Reduviidae (assassin bugs), Pulicidae 
(pulicid fleas), Psychodidae (sand flies), Culicidae (mosquitoes), Muscidae (stable flies), and 
Glossinidae (tsetse flies). This diversity of arthropod species used in the studies indicates that animal 
test systems can be adapted for use with all, or nearly all, species of interest in repellent research and 
development. Of the major families of human-biting arthropods, only the Pediculidae (human lice), 
Ceratopogonidae (biting midges), and Tabanidae (horse flies and deer flies) were not represented in the 
studies reviewed. Eventually, animal test systems may also be developed for use in research and 
development of repellents for use against nonbiting species such as the bush fly, Musca vetustissima 
Walker (Diptera: Muscidae), and the eye gnats, Hippelates colusor (Townsend) and Hippelates pusio 
Loew (Diptera: Chloropidae). 

A total of 21 species of mosquitoes were represented in the studies reviewed, more than half of the 39 
species of all arthropods tested. The number of mosquito species used reflects the generally recognized 
status of the Culicidae as the single most important family of medically important arthropods. Twenty of 


2006 by Taylor & Francis Group, LLC 



128 


Insect Repellents: Principles, Methods, and Uses 


the 53 studies reviewed used the yellow fever mosquito, Aedes aegypti (L.) (Diptera: Culicidae). Its use 
reflects the longstanding status of Aedes aegypti as a preferred species for laboratory studies. 4 ' 5 No other 
arthropod species was used in more than six studies. 


Vertebrate Species 

The first recorded use of animals in biomedical research was by the Greek anatomist Galen (129-200 
ACE). 3 6 According to Locy, 7 the animals used by Galen included Barbary apes (a species of macaque), 
dogs, swine, and cattle. Today the common laboratory animals are rhesus monkeys, dogs, golden 
hamsters, Mongolian gerbils, house mice, Norway rats, guinea pigs, and European rabbits. Most of the 
studies surveyed for the present review used one or more common laboratory species (Table 7.1). This 
circumstance reflects the experience and tradition of the biomedical research community cumulated over 
the course of many years. One of the primary advantages of using the common laboratory species in 
research is the extensive data available on those species in the scientific literature, much of which is 
related in one way or another to the present subject. 

Biomedical research uses the guinea pig so often that the guinea pig has become a metaphor for any 
subject of research, experimentation, or testing. Guinea pigs were used in 18 of the studies surveyed; 17 
studies used mice; and 16 studies used rabbits. Mice have been used in biomedical research for centuries, 
and the mouse has been called “the instrument of biomedical research par excellence.” 8 Besides the 
guinea pig, mouse, and rabbit, no other common laboratory species was used in more than four studies. 
A few studies in the literature surveyed employed poultry and/or livestock as experimental animals. 
Because there are no obvious scientific or technical advantages in using birds and ungulates in lieu of the 
common laboratory species, their use may have been simply a matter of availability. 

Primates are regarded as the most appropriate animals for biomedical research because of their close 
relationship to humans. Because use of the chimpanzee, the closest living relative of humans, is highly 
restricted, the rhesus monkey is the preferred primate for biomedical research. Only two of the studies 
surveyed for the present review employed primates: Kawamura 1 identified the species used in his study 


TABLE 7.1 


Animal Species Used in Research and Development of Repellents for Human Use 


Species 

References 

Birds 


Chicken (domestic fowl) 

2,35 

Pigeon (rock dove) 

42 

Canary 

48 

Mammals 


“Monkey” 

1 

Rhesus monkey 

9 

Dog 

24 

Horse 

48 

Pig 

34 

Ox (cow) 

34,36,49 

Golden hamster 

45 

Mongolian gerbil 

40 

House mouse 

36,45,51,54,67,68,69,70,71,72,74,78,83,90,99,100,101 

Norway rat 

23,37,45,38 

Guinea pig 

1,9,32,33,34,39,41,42,45.48,52,55,58,62,63,72,102,103 

European rabbit 

1,34,38,43,44,45,46,47,50,53,56,59,61,73,75,91 


© 2006 by Taylor & Francis Group, LLC 







Animal Models for Research and Development of Insect Repellents for Human Use 


129 


only as “monkeys,” and, given the time and place of the study, it seems likely that the species used was 
the Japanese macaque, not the rhesus monkey. Coulston and Korte 9 used rhesus monkeys and guinea pigs 
in tests of bicyclic lactones against Aedes aegypti. The relative disuse of rhesus monkeys in repellent 
research and development reflects the high costs of procurement, care, and feeding of primates and the 
cost of the special training required for experimental use of primates. 


Interspecific Differences 

For reasons of economy and human safety, repellents intended for human use are frequently tested on 
a surrogate species, most often the guinea pig, mouse, or rabbit. However, due to specific difference 
between test animals and humans, results obtained in such tests cannot be directly equated to results that 
would be expected in comparable tests on humans. Valid interpretation of results obtained in tests on 
experimental animals depends on the recognition and evaluation of relevant differences between the 
surrogate species and humans. In the case of repellents intended for topical use, the site of interaction of 
the repellent and the target arthropod is the skin. The skin is the largest organ of the body, and its 
structure and function are among the most complex of all organs. It is composed of the epidermis, which 
produces the stratum corneum and the skin pigments, and the dermis, a connective tissue containing 
blood vessels, lymph vessels, nerve endings, the hair follicles, and the skin glands. 

Skin Temperature 

Using data reported by MacNay 10 and Khan et al. 11 Rutledge and Gupta 12 quantified the importance of 
temperature on the effectiveness and persistence of repellents. The MacNay data were collected in field 
tests of repellents against a natural association of Aedes sticticus (Meigen), Aedes stimulans (Walker), 
Aedes vexans (Meigen), and Aedes trichuris (Dyar). Protection periods of the repellents tested decreased 
by an average of 7.6 min for each increase of 1°C in ambient temperature. 12 The data of Khan et al. were 
collected in laboratory tests of deet against Aedes aegypti. The protection period of deet decreased by an 
average of 2.4 min for each increase of 1°C in ambient temperature. 12 

As a rule, skin temperature equilibrates between the body temperature, which is nearly constant in 
healthy individuals, and the ambient temperature, which is variable. Typical skin temperatures of humans 
are in the range 30°C to 32°C, 13 compared with the normal body temperature of 37°C. Normal body 
temperatures of common laboratory animals do not vary more than 3°C from that of humans: rhesus 
monkeys, 38°C; dogs, 39°C; hamsters, 37°C; mice, 35°C; rats, 37°C; guineapigs, 39°C; and rabbits, 40°C. 14 

Skin Permeability 

The skin protects the organism from physical injury and acts as a barrier to the penetration of foreign 
substances. The efficiency of the skin as a barrier to foreign substances depends largely on its thickness. 
The thickness of the skin varies in different species and also on different parts of the body. 15 The skin is 
generally thicker on the dorsal than on the ventral parts of the body, except for the hands and feet, where 
it is thicker on the palms and soles than on the dorsal surface. In humans the thickness of the skin varies 
from about 0.1 mm on the eyelids to as much as 6 mm on the soles of the feet. Human studies have 
resulted in penetration of a topically applied dose of deet into the skin from 9% to 56%. 16 

Blood Content and Circulation 

The network of blood vessels in the skin varies in different species and also on different parts of the body. 15 
The blood content of human skin is greater than the blood content of the skin of any other mammal. 17 


2006 by Taylor & Francis Group, LLC 



130 


Insect Repellents: Principles , Methods, and Uses 


A continuous venous plexus supplied by inflow from the skin capillaries is especially important for the 
supply of blood to the skin. The rate of blood flow into the venous plexus can vary from nearly zero to as 
much as 30% of the total cardiac output. 18 The blood vessels of the dermis affect the effectiveness and 
persistence of repellents applied to the skin by dissipation of body heat through the skin and by the 
absorption and removal of foreign substances from the skin for metabolism and/or excretion elsewhere in 
the body. As indicated above in connection with skin temperature, heat flux at the surface of the skin 
promotes the loss of repellents from the skin by promoting evaporation and convection of fluids at the skin 
surface. In addition, human studies have shown that approximately 17% of a topically applied dose of deet 
is absorbed from the skin into the circulatory system. 16 


Eccrine and Apocrine Sweat Glands 

Active sweat glands wash away foreign substances with sweat. Even minimally functioning sweat glands 
are effective in removing foreign substances from the skin. 19 Two types of sweat glands occur in 
mammals. Eccrine sweat glands secrete a water-and-salt filtrate derived from the blood plasma. Apocrine 
sweat glands secrete organic substances derived from secretory cell cytoplasm in addition to water and 
salt. Eccrine sweat glands predominate in the higher primates, and apocrine sweat glands are restricted to 
small areas such as the armpits and inguinal areas. In lower primates, including the rhesus monkey, and in 
all other mammals, eccrine sweat glands, when present, are restricted to areas of thickened epidermis 
such as the soles; apocrine sweat glands, when present, occur elsewhere on the body. Many species, 
including hamsters, gerbils, mice, rats, guinea pigs, and rabbits have neither eccrine nor apocrine sweat 
glands. 15 


Sebaceous Glands 

The number and size of the sebaceous glands vary in different species and on different parts of the 
body. 15 The sebaceous glands secrete an oily film, sebum, which inhibits penetration of water-soluble 
substances. A variety of lipids, including saturated and unsaturated hydrocarbons and fatty acids, 
compose sebum. In humans, the composition of sebum is known to vary significantly among individuals. 
Correlations have been found between total skin lipid content and the duration of protection of deet and 
between certain fatty acid concentrations and the duration of protection of deet. 20 Studies have also 
shown that a number of skin-surface lipids of humans are repellent to Aedes aegypti. 21 


Hair 

The pelage of mammals varies in different species and on different parts of the body. Humans have 
neither the greatest nor the least amount of hair among mammals. Variably coarse and pigmented 
terminal hairs cover areas of visible hairiness on humans. Seemingly bare areas are covered by fine, 
mostly invisible, vellus hairs, with the exception of on the palms, soles, lips, and a few other small 
areas. 17 As used, topical repellents are applied to exposed, relatively bare areas of skin, excluding the 
scalp where the hair is long. But the skin of most animals is covered with a dense pelage comparable to 
the hair of the human scalp and is therefore not a good model for the relatively bare skin of humans. 
Three methods are used to circumvent this difficulty: (1) infants of species that bear hairless young are 
used in lieu of adult animals^ ’ ; (2) hairless strains or breeds are used in lieu of normal animals 
(hairless strains of dogs, mice, and rats are commercially available); and (3) test animals are shaved on 
the part of the body that is to be treated and exposed to the test insects. Shaving is the most widely 
accepted and frequently used method. 


© 2006 by Taylor & Francis Group, LLC 



Animal Models for Research and Development of Insect Repellents for Human Use 


131 


Technique 

Humane Treatment of Experimental Animals 

The origin of the antivivisectionist movement is associated with the experiments of the French 
physiologists Fran§ois Magendie (1783-1855) and Claude Bernard (1813-1878). 3,6 From France, the 
movement spread to Britain and then to the U.S. The Royal Society for Prevention of Cruelty to Animals 
was founded in 1824. The Cruelty to Animals Act was enacted in 1876, remaining in force until replaced 
with new legislation in 1986. The American Society for Prevention of Cruelty to Animals was founded in 
1866, and the Animal Welfare Act was enacted in 1966 with amendments in 1970 and 1985. The 
touchstone documents of modern animal protection movements are Animal Liberation by Peter Singer 25 
and The Case for Animal Rights by Tom Regan, 26 in which both authors promote a concept of animal 
rights that are collateral to human rights. These works have inspired a large body of literature on animal 
rights and have given rise to a number of organizations promoting this concept, including the radical 
Animal Liberation Front and People for the Ethical Treatment of Animals. 3,27 

Investigators using animal subjects in research must comply with all current laws and regulations 
governing the care and use of experimental animals. The Animal Welfare Act has had the general effect 
of increasing the costs of animal research and reducing the number of experimental animals used in the 
United States. 27 However, requirements for use of animal subjects are generally less burdensome than 
the requirements for use of human subjects, this being a primary advantage of animal research. 

In the U.S., the Animal Welfare Act establishes standards of animal care and use and provides for regular 
inspections by the U.S. Department of Agriculture. Researchers must prepare written protocols for 
approval by an institutional animal care and use committee and consider alternatives to moderate the 
number and/or kind of animals used, including use of in vitro methods in lieu of animals, use of lower rather 
than higher vertebrates, use of advanced statistical techniques to reduce the number of animals needed, and 
refinement of experimental procedures to reduce pain and/or stress. Institutional animal care and use 
committees have power to require changes in protocol to comply with animal care and use standards and to 
stop research projects that do not adhere to an approved protocol. Mroczek 28 has presented a framework for 
understanding pain and suffering in laboratory animals. Also, Toth and Olson 29 have presented strategies 
for minimizing pain and distress in laboratory animals. Two publications useful to researchers at the 
practical level are The Principles of Humane Experimental Technique by Russell and Burch 30 and the 
Guide for the Care and Use of Laboratory Animals published by the National Research Council. 31 

Preparation and Treatment 

Miller and Gibson 32 tested treated netting for irritancy to mosquitoes in a wind tunnel using a guinea pig 
as an attractant without special preparation or treatment. However, most procedures in repellent research 
and development require prior preparation and treatment of the subject. Mechanical restraint may be 
needed to maintain the animal in a position that permits access to the part of the body to which the test 
materials are to be applied and/or to prevent grooming and anti-insect behavior during the procedure. 
Restraining devices include handmade tables, cradles, and boards ~ for restricting movement, and 
hardware cloth, “ sausage casings, or knitted stockinette cloth for close confinement. Restrainers 
for the common laboratory animal species are also available from laboratory equipment and veterinary 
supply firms. Some experimental procedures may require only minor mechanical restraints such as 
collars or stanchions. The procedures of Hill et al. 24 using standing dogs, Kelkar et al. 41 using guinea pigs, 
Rutledge et al. 22 using infant mice, and Mathur et al. 23 using infant rats, did not require use of restraints. 

In practice, mechanical restraint is usually used in combination with chemical restraint (anesthesia). 
With a few exceptions, notably the tabanid flies and the stable fly, Stomoxys calcitrans (L.) (Diptera: 
Muscidae), insect bites are not very painful. The purpose of anesthesia is not only to relieve the discomfort 


2006 by Taylor & Francis Group, LLC 



132 


Insect Repellents: Principles , Methods, and Uses 


and stress of insect bites but also to relieve the discomfort and stress of mechanical restraint. Anesthesia is 
particularly important if the subject is to be restrained for several hours, as in some protocols to determine 
the protection periods of repellents. Some anesthetics frequently used in work with laboratory animals are 
acepromazine, ketamine, pentobarbital, and thiopental sodium. Investigators should consult the insti¬ 
tutional veterinary staff for guidance on the use of anesthetics. 

In escape-box methods ' and olfactometer methods,' the test materials are not usually applied to 
the animal subject, and the shaving and marking of the animal for treatment is not required. But, except in 
the case of hairless animals, it is usually necessary to shave the part of the body to which the test 
materials will be applied. Then, except in the case of systemic treatments 41 ' 44 " 45 or of whole-body 
treatments such as dips " and sprays, ’ a template, cutout, or pen is used to enclose or mark off the 
shaved treatment area(s). 24,39,46,47 Test materials are applied generally to the treatment area(s) by pipette 
as ethanol solutions in concentrations and volumes calculated to provide the desired application rates in 
mg/cm 2 or pg/cm 2 of skin. For planning purposes, the maximum rate of application of a liquid that can be 
accomplished without runoff is about 2 mg/cm 2 . Fully formulated products are applied undiluted at the 
rate specified in the "Directions for Use” section of the product label. 


Test Methods 

In four of the studies reviewed, the treated animals were exposed to natural populations of insects in the 
held rather than to caged insects in the laboratory. 1,37,48,49 Field studies have the advantage that the 
environmental conditions of the study are closer to those under which the end-use product will be used, 
while laboratory studies have the advantage that the conditions of the study can be more closely 
controlled to reduce experimental error. 

In small-cage methods of exposure, a cage containing a known number of test insects is applied to the 
treated part of the animal’s body. The test insects are allowed access to the treatment(s) by withdrawing a 
slide or by some other means. 47,50 In large-cage methods, one or more treated animals are placed inside 
a cage containing a known number of test insects. 51,52 Except in the case of whole-body treatments, 
untreated parts of the animal's body are excluded from the test insects. The experimental design may also 
be such that each insect test population has access to one treatment only (no-choice method of 
exposure) 53,54 or that each insect test population has access to any of two or more treatments (free- 
choice method). 55,56 Curtis et al. 57 have shown that no-choice methods and free-choice methods do not 
provide equivalent results. 

Test Population Size 

Although natural insect population densities vary widely, densities of insect test populations are normally 
standardized in laboratory studies. The procedure in which the behavior of individual animals is observed 
and recorded is called focal sampling. For example. Miller and Gibson, " Dethier, and Galun et al. 
observed and recorded the responses of individual mosquitoes and tsetse flies to various test materials. 
Where larger numbers have been used, there has been little or no coordination or agreement among 
investigators on test population size or test cage size. For 17 reports in which insect test population sizes 
were stated for the small-cage method, the range was from 5-10 61 to 2,000, 47 and the median was 20. For 
13 reports in which insect test population size was stated for the large-cage method, the range was from 
20 36,38,51 to 1,000-2,000 40 and the median was 94. 

Similarly, for 15 reports in which test cage shape and size were stated for the small-cage method, 
the range of the computed volume was from 5.0 X 10 cnr to 1.6 X 10 cm , and 1.6 X 10" cm was the 
median. For 11 reports in which test cage shape and size were stated for the large-cage method, 

3 3 7 3 4 3 

the range of the computed volume was from 1.3X10' cm' to 1.6X10 cm , and 3.0X10 cm was 
the median. 36 The value 1.6 X 10 7 cm 3 was not representative of the set; it refers to a large cage made 
to enclose cattle into which 100 Stomoxys calcitrans were released. 


© 2006 by Taylor & Francis Group, LLC 



Animal Models for Research and Development of Insect Repellents for Human Use 


133 


For 12 reports in which both insect test population size and test cage shape and size were stated for the 
small-cage method, the range of the computed insect population density was from 1.9 insects per 
1,000 cm 3 to 5.6X 10 2 insects per 1,000 cm 3 , and 5.9X 10 1 insects per 1,000 cm 3 was the median. 43 ’ 63 
For eight reports in which both insect test population size and test cage shape and size were stated for the 
large-cage method, the range of the computed insect population density was from 6.2 X 10 _3 insects per 
1,000 cm 3 to 2.6X 10 2 insects per 1,000 cm 3 , and 3.6 insects per 1,000 cm 3 was the median. 36 ' 40 The 
value 6.2X10 -3 insects per 1,000 cm 3 was not representative of the set, referring to the experiment 
described above involving a large cage made to enclose cattle. 

Considerations of economy in rearing and handling of test insects and considerations of animal welfare 
favor the use of small numbers of insects, while considerations of statistical precision favor the use of large 
numbers of insects. Khan et al. 64 and Barnard et al. 65 have shown that insect test populations of different 
density do not provide equivalent estimates of protection time when the test method used depends on a 
fixed endpoint for protection time such as the first (or second) observed bite. This is because the first (or 
second) insect to bite in a small test population represents a less extreme position in the tolerance 
distribution than the first (or second) insect to bite in a large test population. 66 


Observing and Recording Test Data 
Methods of Observing and Recording 

Preliminary, informal observations aimed at clarifying and finalizing details of technique, experi¬ 
mental design, and data analysis are almost always necessary in repellent research, because the test 
insects, test subjects, materials, equipment, facilities, and personnel involved form a complex and 
variable system that may not provide accurate and definitive results as anticipated in planning the 
study. In experiments on animals, scoring is usually based on observations of biting, full or partial 
feeding, or attachment of the test arthropods. Most of the studies that were reviewed relied on visual 
observation of the data to be recorded. However, Lai et al. 67 demonstrated radioactive tracer and 
fluorescent dye techniques in tests of repellents on mice against Aedes aegypti. Also Kashin and 
Kardatzke 68 ’ 69 and Kashin and Arneson 70 electronically recorded the time of each bite in tests of 
repellents on mice against Aedes aegypti. 

In the terminology of behavior studies, 60 procedures for recording bites in a repellent test or experiment 
may be either continuous recording, meaning that the observer records the occurrence and time of each bite 
from the beginning to the end of the procedure, or time sampling, meaning that the observer records the 
occurrence of biting periodically during the procedure (for examples of continuous recording, see Kashin 
and Arneson, 68 Kashin and Kardatzke, 69 ’ 70 Sachdeva et al. 71 and Abu-Shady et al. 72 ). 

Time sampling may be either instantaneous sampling or one-zero sampling. In instantaneous sampling 
the observation session is divided into successive periods of time called sample intervals. The instant of 
time at the end of each sample interval is called a sample point, and the observer records the biting 
activity of the test insects “instantaneously” at each sample point (for examples of instantaneous 
sampling, see Wirtz et al. 73 and Choi et al. 74 ). In one-zero sampling, the observation session is similarly 
divided into sample intervals, but at each sample point the observer records all bites that have occurred 
during the preceding sample interval. In repellent studies, there may be only one sample interval that 
extends from the beginning to the end of the observation session, and the number of insects that have fed 
on (argasid ticks, reduviid bugs, mosquitoes, biting flies, fleas) or attached to (mites, ixodid ticks, 
sticktight fleas) the animal subject is determined at the end (for examples of one-zero sampling see 
Tripathi et al. 43 and Fryauff et al. 61 ). 

Recording Protection Time 

Both continuous and time sampling methods are used in tests to determine the protection times of 
repellents. In a common procedure the treatment is exposed to the test insects continuously or at intervals 


2006 by Taylor & Francis Group, LLC 



134 


Insect Repellents: Principles , Methods, and Uses 


until a specified endpoint such as the first (or second) observed bite is reached. In an alternative procedure 
demonstrated by Hill et al. 24 in tests of repellents on dogs, the treatment is applied at intervals to different 
test subjects or at intervals to different treatment areas on the same subject. All subjects or areas are 
exposed subsequently to the test insects at the same time, and the number of bites received is recorded at 
that time. 

Researchers have overestimated consistently protection times determined by time sampling to a fixed 
endpoint such as the first (or second) observed bite because protection times have been recorded 
traditionally as the period from the time of application of the treatment to the time of the sample point at 
which the first (or second) bite is observed, ignoring the possibility that the first (or second) bite might 
have occurred at another time during the preceding sample interval if the test subject had been available 
(for examples of sample interval error, see Bar-Zeev and Ben-Tamar 75 and Bar-Zeev and Gothilf 62 ). 
Kasman et al. 39 reported a mathematical correction for sample interval error. Rutledge 76 recommended 
recording the midpoint of the sample interval preceding the sample point at which the endpoint of the test 
was observed. 


Experimental Design and Data Analysis 
Null Treatment (Control Experiment) 

The use of controls in biological experiments is universally accepted, 77 but some investigators using 
human subjects in repellent studies have limited the number of bites received by control subjects: (1) by 
restricting the control subjects’ exposure to a smaller area of skin than the treated subjects; (2) by 
restricting the control subjects’ exposure to the beginning and end of the test only or to short periods of 
time during the test; (3) by preventing actual bites and counting landings instead; (4) by substituting a 
standard treatment for the null treatment; or (5) by dispensing with the null treatment altogether. Such 
shortcuts and substitutions inevitably lead to ambiguity in the data obtained because the biting activity of 
the insect test population is in a constant state of flux in the course of any repellent test. A bona fide 
control experiment, defined as an experiment that duplicates the primary experiment in every way except 
for inclusion of the test material, is an essential part of any repellent test procedure. In this regard, the use 
of animal subjects offers a distinct advantage over the use of human subjects, who are not anesthetized 
during repellent test procedures. 

Protection Time Models 

Traditionally, estimates of protection time have been assumed to be random variables from a normal 
distribution for purposes of analysis. The purpose of this section is to point out some implications of this 
assumption and to suggest some alternative models used in science and technology to analyze 
analogous data. 

Protection Time as a Random Variable from a Normal Distribution 

As stated above in connection with insect test population size, the first (or second) individual to bite in 
a population of insects is the individual that occupies the most (or next to most) extreme position in the 
tolerance distribution of that population, i.e., the individual whose position in the tolerance distribution is 
most (or next to most) distant from the mean of the test population. It is, in other words, the individual 
that is the least (or the next least) representative of the population as a whole with respect to tolerance for 
the test material. Statistically, the consequence of using the first or second bite as the endpoint for 
determination of protection time is that the variance of the estimate obtained is large compared with the 


© 2006 by Taylor & Francis Group, LLC 



Animal Models for Research and Development of Insect Repellents for Human Use 


135 


variance of estimates obtained from observations of less extreme individuals. Rutledge et al. 66 have 
demonstrated this point in terms of the standard normal distribution. 

Observations from Truncated and Censored Distributions 

Traditional protection time methods, such as the method used by Smith et al. 34 in tests of repellents 
on guinea pigs, rabbits, swine, and cattle against Aedes aegypti and the method used by Li et al. 78 in 
tests of repellents on mice against Aedes aegypti, depend on a fixed endpoint such as the first (or 
second) observed bite. In other words, the protection time of the test material is observed for only 
that individual that is the first to bite (or for only those individuals that are the first and second to 
bite). The protection time of the test material is not determined for any other member of the insect 
test population. Samples obtained when observation is restricted over a portion of the sample space 
are known as truncated and censored samples. 79 In terms of protection time, truncated samples are 
those samples in which the number of test insects with protection times that lie within the restricted 
area is not known, as in a field test; censored samples are those samples in which the number of test 
insects with protection times that lie within the restricted area is known, as in a laboratory test. 
Because all the protection times that are not observed in the test are known to be longer than those 
for the first and second individuals to bite, the samples are said to be right singly truncated or right 
singly censored. Because the values of protection time obtained in this kind of test method are 
extreme values, the extreme value distribution may be more appropriate than the normal distribution. 
Cohen 79 provides examples of the analysis of truncated and censored samples from both the normal 
and the extreme value distribution. 


Failure Time Data 

Data obtained in tests in which the time to occurrence of a specified event is observed and recorded 
are called failure time data. 80 In failure time terminology, a bite recorded in a protection time test, 
such as the first or second observed bite, is regarded as a “failure” with respect to the test material, 
and the time to occurrence of the bite is called the failure time. As stated in the preceding section, 
data obtained in protection time procedures that depend on a fixed endpoint such as the first (or 
second) observed bite are truncated or censored. In failure time terminology, truncation or censoring 
is called order statistic, or type II, truncation or censoring because the test is discontinued when the 
first (or second bite) has been observed. Kalbfleisch and Prentice 80 discuss the analysis of failure time 
data in terms of various distributions, including the Weibull distribution. The Weibull distribution is 
related to the extreme value distribution, and results obtained in terms of either can be transferred to 
the other. 

Singularities in Catastrophe Data 

Mathematically, a sudden change caused by gradual alteration of circumstance is termed 
a catastrophe. Accordingly, the occurrence of the first, second, and succeeding bites in the course 
of a protection time test can be interpreted in terms of catastrophe theory. In this interpretation, the 
successive bites inflicted by members of the insect test population are catastrophic events, or 
singularities, induced by smooth changes in the various factors modulating biting behavior in the test 
environment, of which the factor of primary interest and importance is the test material itself. It has 
been demonstrated that the gradual loss of effectiveness of a repellent on the skin conforms to the 
half-life law. 81 When the deposit has dissipated to the level of tolerance of the most tolerant 
individual in the insect test population, the first bite will occur, when the deposit has dissipated to the 
level of tolerance of the next most tolerant individual in the insect test population, the second bite 
will occur, and so on for succeeding bites. Poston and Stewart 82 give an integrated treatment of the 
main ideas of catastrophe theory. 


2006 by Taylor & Francis Group, LLC 



136 


Insect Repellents: Principles, Methods, and Uses 


Analysis of Variance 

This section addresses the experimental designs for analysis of variance that are or have been in common 
use in repellent research and development. 

Paired Observations 

The familiar t test has been applied in a variety of repellent studies. For example. Wood 55 determined the 
relative effectiveness of ethyl hexanediol at 35% and 82% RH in paired observations using guinea pigs 
and Aedes aegypti. Similarly, Shirai et al. 83 determined the relative effectiveness of various concen¬ 
trations of L-lactic acid with null treatments in paired observations using mice and Aedes albopictus 
(Skuse) (Diptera: Culicidae). The method of paired observations is also used to compare the effectiveness 
and/or persistence of new or candidate repellents with that of a material standard. 

One-Way and Multi-Way Experimental Designs 

One-way, or completely random, designs are relatively simple to design, execute, analyze, and interpret. 
The variance components are those for treatments, error, and sampling units, if subsampling is included 
in the experimental design. For example, Miller and Gibson 32 used the one-way analysis of variance to 
analyze responses of Anopheles gambiae Giles and Culex quinquefasciatus Say (Diptera: Culicidae) 
to permethrin, pirimiphosmethyl, and lambdacyhalothrin in a wind tunnel baited with a guinea pig. 

Multi-way designs include randomized complete block, Latin square, factorial, splitplot, and many 
other designs. Such designs, in which the observations are cross-classified by blocks, plots, factors, etc., 
are used to reduce the variance of treatment means and to increase the scope of inference of the 

Q/1 oc o<t , , t 

experiment. For example. Fryer et al. ’ used multi-way analysis of variance designs in tests of fly 
repellents on cattle. 

Balanced Incomplete Block Designs 

In 1945, F.A. Morton introduced the balanced incomplete block design into repellent research and 
development as a way to reduce the variance of treatment means by segregating the variance attributable 
to differences among test subjects. 87 Ten years later, Altman and Smith 88 introduced a mathematical 
formula for computing adjusted treatment means in balanced incomplete block design experiments. This 
formula has been widely used in repellent research and development since 1955 and has been included in 
a standard method published by the American Society for Testing and Materials. 89 However, the formula 
of Altman and Smith 88 is erroneous, and treatment means computed with it are inaccurate. 76 In some 
cases, adjusted treatment means computed with this formula do not lie within the range of observed 
values for the treatment. Additionally, in some cases the adjusted treatment mean may even be negative. 
For example, in Table 1, a report of tests of repellents on rabbits against Ornithodoros tholozani 
Laboulbene & Megnin (Acari: Argasidae), Bar-Zeev and Gothilf 62 gave the adjusted mean protection 
time of compound 14458 as 2.44 h and the range of observed protection times as 1-2 h. Obviously, use of 
the formula of Altman and Smith for adjusting treatment means of balanced incomplete block 
experiments is misuse of the balanced incomplete block design. 

Bioassay Methods 

In bioassay test methods the responses of the test population to the test material are determined over 
a range of doses (application rates). The results of testing are analyzed as the linear regression of response 
(probit transformation) on dose (logarithmic transformation) (Figure 7.2) to obtain estimates of the 
median effective dose (ED 50 ), the 95% or 99% effective dose (ED 95 or ED 99 ), and their respective 
confidence limits. The basic method of bioassay and its variations are indispensable in modern science 
and technology, particularly in the fields of pharmacology and toxicology, including insect toxicology. 


© 2006 by Taylor & Francis Group, LLC 


137 

Probit % 

- 7.5 99.38 

- 7.0 97.7 

- 6.5 93.3 

- 6.0 84.1 

- 5.5 69.1 

- 5.0 50.0 

- 4.5 30.9 

- 4.0 15.9 

- 3.5 6.7 

. 3.0 2.3 

0.4 0.6 0.8 1.0 1.2 1.4 1.6 

Log concentration 

FIGURE 7.2 Illustrating the probit transformation: The sigmoid curve of per cent response (vertical axis) on repellent 
concentration (horizontal axis) becomes linear when the percent response is transformed to the probit scale and the repellent 
concentration is transformed to the logarithmic scale. (From D. J. Finney, Probit Analysis , 3rd ed., London: Cambridge 
University Press, 1971.) 

No-Choice and Free-Choice Designs 

The statistical methods employed in bioassay differ in no-choice and free-choice experimental designs. 
The data obtained in a no-choice test are all-or-nothing, or quantal, data. Specifically, the test insects in a 
no-choice repellent test may either feed or not feed. There is no other alternative. On the other hand, data 
obtained in free-choice repellent tests are quantitative, or nonquantal, data because a test insect may either 
feed or not feed on any of two or more alternative treatments. In addition, free-choice repellent tests usually 
employ an instantaneous sampling method because it is usually not possible or practicable to determine on 
which treatment each insect has fed when the test is terminated. For example, Yeoman et al. 90 tested butyl 
3-methylcinchoninate on mice against Stomoxys calcitrans and Galun et al. 3 tested microencapsulated 
pyrethrum on guinea pigs against Glossina morsitans Westwood (Diptera: Muscidae) and Ornithodoros 
tholozani by the no-choice repellent bioassay method. Robert et al. 91 tested five repellents on rabbits 
against four species of Anopheles. Choi et al. 74 tested several repellents on mice against Culex pipiens L. 
(Diptera: Culicidae) by the free-choice repellent bioassay method. 

Probit Analysis by D.J. Finney 92 is the classical reference on the statistical methods of bioassay. 
Earlier editions of this book included the method for probit analysis of quantitative data, but these were 
eliminated in the third (1971) edition because, according to the author, “The problem is not very 
common.” The method for probit analysis of quantitative data is also available in Goldstein. 93 However, 
the no-choice method is heavily favored in research and development and is the standard laboratory 
method in nearly all insecticide studies. Curtis et al. 57 have demonstrated use of the logit transformation. 



2006 by Taylor & Francis Group, LLC 



138 


Insect Repellents: Principles, Methods, and Uses 


which is based on the logistic (autocatalytic) growth function, in lieu of the probit transformation, which 
is based on the binomial distribution of probabilities, in the analysis of data obtained in free-choice 
bioassays. Nonetheless, it is known that there is no practical difference in results of analyses using the 
logit and probit transformations. 94 

Effective Dose 

Although the effective dose of the test material can be computed from bioassay data for any desired 
fraction of the insect test population except 0% and 100%, the doses normally computed are the ED 50 , for 
statistical use, and the ED 95 or ED 99 , for practical use. Unless otherwise indicated, the term “effective 
dose” is understood to mean the dose that is effective at the stated level (50%, 95%, or 99%) at the time of 
application. An effective dose determined for a longer period (for example, 4 h after application) is so 
designated (for example, the 4 h effective dose). A standard method for determining the effective dose 
and the 4 h effective dose of repellents on humans against mosquitoes has been published by the 
American Society for Testing and Materials. 95 

Smith et al. 34 defined the “minimum effective dosage” as “the minimum amount [of the test material] 
per unit of surface required to protect against the given population of insects.” That would be the 100% 
effective dose or ED 10 o, and, as Finney 92 has stated, “Even a very large experiment could scarcely 
estimate [the ED 10 o] with any accuracy.” An additional source of confusion is that the abbreviation MED 
has been used variously to designate the median effective dose, minimal effective dose (terminology of 
Finney 92 ), and “minimum effective dosage” (terminology of Smith et al. 34 ). 


Persistence 

Two primary attributes of a topical repellent are its effectiveness, i.e., its ED 50 and its ED 95 or ED 99 , 
and its persistence on the skin. 81 Traditionally, persistence has been defined in terms of protection 
time, i.e., the time elapsing from the time application of the test material to the time at which the first 
(or second) bite is obtained from the insect test population. When defined in this way, the observed 
protection time varies with the dose applied and the density of the insect test population. Bioassay 
methods minimize this uncertainty by testing a range of doses over time and by substituting a 
proportional endpoint (95% or 99% effectiveness) for the absolute endpoint (the first or second bite) 
traditionally employed. 66 ' 81 

Probit Plane Model. In the probit plane method for bioassay of repellents, the number of bites 
permitted by each of several doses of the test material (for example, 0.0, 0.2, 0.4, 0.8, and 1.6 mg/cm 2 ) is 
determined at each of several different times after application (for example, at 0, 2, 4, 6, and 8 h). The 
data obtained are analyzed as the multiple regression of response (probit transformation) on dose 
(logarithmic transformation) and time (Figure 7.3). Estimates provided by the multiple regression 
equation include the median, 95%, and 99% protection time for any desired dose within the range of 
doses tested; the median, 95%, and 99% effective dose for any desired time within the range of times 
tested; and the confidence limits these estimates at any desired level of confidence. The basic reference 
on probit plane bioassay is that of Finney. 92 Rutledge et al. 66 ’ 81 ’ 96 have demonstrated the probit plane 
method in tests of deet and ethyl hexanediol on humans against Aedes aegypti in the laboratory and Aedes 
dorsalis (Meigen) in the field. 

Effective Half-Life. Rutledge et al. 81 have suggested that the effective half-life could be an 
alternative to protection time as a measure of repellent persistence. Computation of effective half-life 
was demonstrated with data obtained in probit plane bioassays of deet and ethyl hexanediol on 
humans against Aedes aegypti (Figure 7.4). When the effective halflife and the effective dose of a 
repellent are known, it is possible to estimate the initial dose required to provide a given level of 
protection (for example, 95%) for a given time (for example, 4 h) after application and to estimate the 
time that a given initial dose (for example, 2 mg/cm 2 ) will remain effective at a given level of 
protection (for example, 95%). 


© 2006 by Taylor & Francis Group, LLC 


Animal Models for Research and Development of Insect Repellents for Human Use 


139 


y (+1,0,10.39) 



FIGURE 7.3 Illustrating the probit plane model: The response of Aedes aegypti to deet (Y axis, in probit values) is 
represented as a plane plotted on the applied dose (X! axis, logarithmic scale) and the elapsed time from the time of 
application (X 2 axis, hours). Dark circles and associated vertical lines show the observed values and their respective 
deviations from the probit plane. (From L. C. Rutledge et al.. Mathematical models of the effectiveness and persistence of 
mosquito repellents, J. Am. Mosq. Control Assoc., 1, 56, 1985.) 


Extrapolation to Humans 

Material Standards and Comparative Observations 

A material standard is a standardized material to which other materials can be compared in paired 
observations. For example, Yeoman et al. 90 determined the effectiveness of six doses of butyl 
3-methylcinchoninate (the test material) and deet (the material standard) against Stomoxys calcitrans 
in paired observations on mice. The ED 50 ’s of the test material (0.002 mg/cm 2 ) and material standard 
(0.01 mg/cm 2 ) were estimated graphically. If the ED 50 of the material standard in comparable tests on 
humans were known, and if the difference between the ED 50 ’s of test materials and the material standard 
( — 0.008 mg/cm 2 ) were known to be the same in comparable tests on mice and humans, then an 


© 2006 by Taylor & Francis Group, LLC 









140 


Insect Repellents: Principles, Methods, and Uses 



FIGURE 7.4 Illustrating the half-life of deet on the human forearm: The applied dose was found to decay at a rate (A) of 
1.04 log mg/cm 2 per hour, from which the half-life (ty 2 ) was computed to be 0.67 hours. Xi is the applied dose (0.2 mg/cm 2 ) 
and Z is the residue remaining at X 2 hours after application. The dashed lines indicate the computed ED 50 and ED 95 . (From 
Letterman Army Institute of Research, San Francisco, CA.) 


extrapolation from the value obtained in tests on mice to the value that would be expected in tests on 
humans could be made (See section on correction terms). 

Similarly, Bar-Zeev and Gothilf 63 determined the protection times of 538 organic compounds (the test 
materials) and deet (the material standard) against Xenopsylla cheopis (Rothschild) (Siphonaptera: 
Pulicidae) in paired observations on guinea pigs. The ratios of the mean protection times of the test 
materials to the respective mean protection times of deet were computed for purposes of comparison. If 
the protection time of the material standard in comparable tests on humans were known, and if the ratios of 
the protection times of the test materials to the material standard were known to be the same in comparable 
tests on guinea pigs and humans, then extrapolations from values obtained in tests on guinea pigs to values 
that would be expected in tests on humans could be made (See section on correction factors). 

Statistical Adjustment of Data 

Because animal models do not precisely simulate the human standard, it is necessary to adjust values 
obtained in experiments on animals statistically to estimate the corresponding values that would be 
obtained in comparable experiments on humans. In this context, an adjusted value is defined as a derived 
value that can be used for an intended purpose. 97 The problem of adjusting experimental data occurs in 
many different fields of science and technology and particularly in the fields of pharmacology 
and toxicology. 

Correction Terms 

In the present context a correction term, or additive correction, is a derived value that can be added to a 
value obtained in an animal test system to estimate the corresponding value that would be obtained in 
a comparable human test system. The correction term is computed as the mean difference between values 
obtained in the animal and the human test system, using data obtained by testing a series of materials in 


© 2006 by Taylor & Francis Group, LLC 





Animal Models for Research and Development of Insect Repellents for Human Use 


141 


both systems. Corrected values of estimates subsequently obtained in the animal test system, i.e., 
adjusted values that estimate corresponding values in the human test system, are computed by adding the 
correction term to estimates obtained in the animal test system. The variance of a corrected value is equal 
to the variance of the value obtained in the animal test system plus the variance of the correction term. 

For example, Rutledge et al. 98 tested eight repellents on mice and humans against Aedes aegypti using 
the probit plane method. Correction terms for the Y intercept and each of the two regression coefficients 
were computed from data obtained on four of the repellents and verified with data obtained on the other 
four. Corrected values computed from tests on mice did not differ significantly from values obtained in 
tests on humans. Similarly, Kasman et al. 39 determined graphically that the protection times of a number 
of repellents obtained in tests on humans against Aedes aegypti were approximately 20 min longer than 
those obtained in tests on guinea pigs. In this study, then, the value of the correction term was +20 min. 

Correction Factors 

In the present context a correction factor, or multiplicative correction, is a derived value that can be 
multiplied by a value obtained in an animal test system to estimate the corresponding value that would be 
obtained in a comparable human test system. The correction factor is computed as the mean ratio of the 
values obtained in the animal and the human test system, using data obtained by testing a series of 
materials in both systems. Corrected values of estimates subsequently obtained in the animal test system, 
i.e., adjusted values that estimate corresponding values in the human test system, are computed by 
multiplying estimates obtained in the animal test system by the correction factor. The variance of a 
corrected value is equal to the variance of the value obtained in the animal test system times its mean plus 
the variance of the correction term times its mean, if the variances are small compared to the means. 

No exact example of the use of correction factors was found in the literature reviewed. However, the 
practice of converting animal test data to ratios of the value obtained on the test material to the value 
obtained on a material standard is suggestive of the correction factor approach. Also Hill et al. 24 
computed the linear regression of the protection time of repellents on guinea pigs against Aedes aegypti 
on the protection time of the same repellents on humans against Aedes aegypti. In this case, an additive 
correction (the Y intercept) and a multiplicative correction (the regression coefficient) were used in 
conjunction. In this study, however, the regression line fitted was inverted (i.e., guinea pig values were 
the dependent variable). Values of the Y intercept and the regression coefficient were not reported. 

Curve Fitting 

In the graphical method of curve fitting, values obtained in the human test system (the dependent 
variable) are plotted against values obtained in the animal test system (the independent variable), using 
data obtained by testing a series of materials in both systems. Values subsequently obtained in the animal 
test system can then be converted to estimates of values to be expected in the human test system by 
reading the latter from the graph. 

Statistical methods of curve fitting are more precise and provide confidence limits for the values 
estimated. In the study of Hill et al. 24 the relation between protection times of repellents on guinea pigs 
and humans against Aedes aegypti was linear. If the relation between the variables studied is nonlinear, 
the data may be either transformed to yield a linear relation or analyzed by the methods of curvilinear 

84 

regression. 


Conclusion 

History shows that nearly all significant advances in biomedical science are made first in experiments on 
animals. Some familiar examples are the development of physiology and biochemistry through 


2006 by Taylor & Francis Group, LLC 



142 


Insect Repellents: Principles, Methods, and Uses 


experiments done primarily on mice and rats, toxicology and endocrinology in experiments done on 
rabbits, and genetics in experiments done on fruit flies. The twentieth century advances in pharmacology 
and toxicology can be regarded as prototypical of the advances in repellent science currently in progress 
because the basic problem of quantifying and measuring the response of the organism to the test material 
is the same in each case. Because no animal test system for repellents has come into general use to date, 
the researchers cited in the present paper are best regarded as pioneers of a methodology that has not yet 
reached maturity. Collectively, however, they have demonstrated the principles and procedures that will 
shape the mature methodology that is eventually standardized and adopted for general use. 

The standardization and general adoption of animal repellent test systems will depend on the 
development and refinement of available techniques for increased precision and accuracy and for 
accurate extrapolation to comparable human test systems. However, the twin problems of precision 
and accuracy 104 apply equally to both animal and human test systems, and the problem of accurate 
extrapolation from one system to the other can be resolved only by improving the precision and 
accuracy of both. Although animal test systems have been developed, or invented, entirely on an ad 
hoc basis to date, standardization of animal test systems for general use will require funding of a 
research and development project, including a program of interlaboratory trials, dedicated to that 
specific end. In computing cost-benefit figures for the project, account should be taken not only of the 
potential savings in the long-term costs of repellent research and development but also the projected 
increase in human safety resulting from deferral of tests on humans to the late stages of 
repellent development. 

The potential of experiments on animals in repellent science is even greater in the area of basic 
research than it is in the area of applied research. Chemoreception, structure-activity relationships, mode 
of action, and other basics of biochemistry, physiology, and ethology bearing on the interaction of 
repellent, arthropod, vertebrate host, and environment are still little known. Basic principles of 
combining and formulating repellent compounds for controlled release, reduced absorption, abrasion 
resistance, synergism, user acceptance, and other desirable properties are similarly little known. 
Advances in repellent science can not be achieved through the development of precise, accurate test 
methods alone. Advances in basic knowledge also are needed, and, as the history of biomedical science 
has shown, the most productive approach to advances in basic knowledge is that of basic research 
on animals. 


References 

1. R. Kawamura, Studies on tsutsugamushi disease (Japanese flood fever), Med. Bull. Coll. Med. Univ. 
Cincinnati, 4(Suppl.), 1, 1926. 

2. A. W. Lindquist, A. H. Madden, and C. N. Watts, The use of repellents against fleas, J. Econ. 
EntomoL, 37, 485, 1944. 

3. A. Guerrini, Experimenting with Humans and Animals: From Galen to Animal Rights. Baltimore: The 
Johns Hopkins University Press, 2003. 

4. S. R. Christophers, Aedes aegypti (L.), the Yellow Fever Mosquito: Its Life History, Bionomics and 
Structure, London: Cambridge University Press, 1960. 

5. G. B. Craig, The contribution of Aedes aegypti research to the advancement of biological science. Am. 
J. Trop. Med. Hyg., 14, 904, 1965. 

6. E. Gasking, The Rise of Experimental Biology, New York: Random House, 1970. 

7. W. A. Locy, The Story of Biology, Garden City. NY: Garden City, 1925. 

8. R. Bud and D. J. Warner (Eds.), Instruments of Science: An Historical Encyclopedia, London: The 
Science Museum, 1998. 

9. Coulston, F. and Korte. F. W. A. G. K., Insect Repellent, U.S. Patent 4,663.346, 1987. 

10. C. G. MacNay, Studies on repellents for biting flies, Can. Entomol., 71, 38, 1939. 

11. A. A. Khan, H. I. Maibach, and D. L. Skidmore, A study of insect repellents, 2. Effect of temperature 
on protection time, J. Econ. Entomol., 66, 437, 1973. 


© 2006 by Taylor & Francis Group, LLC 



Animal Models for Research and Development of Insect Repellents for Human Use 


143 


12. L. C. Rutledge and R. K. Gupta, Reanalysis of the C.G. MacNay mosquito repellent data, J. Vector 
Ecol., 21, 132, 1996. 

13. B. G. J. Knols, W. Takken, and R. De Jong, Influence of human breath on selection of biting sites by 
Anopheles albimanus, J. Am. Mosq. Control Assoc., 10, 423, 1994. 

14. W. S. Spector (Ed.), Handbook of Biological Data, Philadelphia: W.B. Saunders, 1956. 

15. V. E. Sokolov, Mammal Skin, Berkeley: University of California Press, 1982. 

16. P. J. Robbins and M. G. Cherniak, Review of the biodistribution and toxicity of the insect repellent 
AUV-diethyl-zn-toluamide (deet), J. Toxicol. Environ. Health, 18, 503, 1986. 

17. W. Montagna, The Structure and Function of Skin, 3rd ed., San Diego: Academic Press, 1974. 

18. A. C. Guyton and J. E. Hall, Human Physiology and Mechanisms of Disease, 6th ed., Philadelphia: 
W.B. Saunders, 1997. 

19. R. L. Rietschel and T. S. Spencer, Correlation between mosquito repellent protection time and 
insensible water loss, J. Invest. Dermatol., 65, 385, 1975. 

20. W. A. Skinner et al.. Influence of human skin surface lipids on protection time of topical mosquito 
repellent, J. Pharm. Sci., 66, 1764, 1977. 

21. W. A. Skinner et al.. Human skin-surface lipid fatty acids—mosquito repellents, Experientia, 26, 728, 
1970. 

22. L. C. Rutledge, M. A. Lawson, and L. L. Young, Tests of repellents against Diamanus montanus 
(Siphonaptera: Ceratophyllidae), J. Med. Entomol., 19, 361, 1982. 

23. K. K. Mathur et al., Laboratory evaluation of diethyl phenyl acetamide & other insect repellents 
against rat flea, Xenopsylla cheopis (Rothschild), Indian J. Med. Res., 83, 466, 1986. 

24. J. A. Hill et al., Evaluation of mosquito repellents on the hairless dog, Mosq. News, 39, 307, 1979. 

25. P. Singer, Animal Liberation: A New Ethics for Our Treatment of Animals, New York: New York 
Review of Books, 1975. 

26. T. Regan, The Case for Animal Rights, Berkeley: University of California Press, 1983. 

27. M. Mukerjee, Trends in animal research, Sci. Am., 276(2), 86, 1997. 

28. N. S. Mroczek, Recognizing animal suffering and pain. Lab Animal., 21, 27, 1992. 

29. L. A. Toth and G. A. Olson, Strategies for minimizing pain and distress in laboratory animals. Lab 
Animal, 20, 33, 1991. 

30. W. L. M. Russell and R. L. Burch, The Principles of Humane Experimental Technique, London: 
Methuen, 1959. 

31. National Research Council, Guide for the Care and Use of Laboratory Animals, Washington, DC: 
National Academies Press, 1996. 

32. J. E. Miller and G. Gibson, Behavioral response of host-seeking mosquitoes (Diptera: Culicidae) to 
insecticide-impregnated bed netting: A new approach to insecticide bioassays, J. Med. Entomol, 31, 
114, 1994. 

33. R. Galun et ah. Long-term protection of animals from tsetse bites through controlled-release 
repellents, Isotope and Radiation Research on Animal Diseases and their Vectors, Vienna: 
International Atomic Energy Agency, 1980, p. 207. 

34. C. N. Smith et ah. Factors affecting the protection period of repellents, U.S. Dept. Agric. Tech. Bull., 
1285, 1963. 

35. K. Watanabe et ah, New mosquito repellent from Eucalyptus camaldulensis, J. Agric. Food Chem., 41, 
2164, 1993. 

36. G. W. Eddy and W. S. McGregor, Use of white mice for testing materials used as repellents and 
toxicants for stable flies, J. Econ. Entomol, 42, 461, 1949. 

37. M. Bar-Zeev and S. Gothilf, Field evaluation of repellents against the tick Ortithodoros tholozani 
Labou & Megn., J. Med. Entomol., 11, 389, 1974. 

38. M. Bar-Zeev et ah. Studies of repellents against Panstrongylus megistus (Hemiptera: Reduviidae), 
J. Med. Entomol., 1, 70, 1980. 

39. S. Kasman, L. A. O. Roadhouse, and G. F. Wright, Studies in testing insect repellents, Mosq. News, 13, 
116, 1953. 

40. J. A. Klocke, M. V. Darlington, and M. F. Balandrin, 1, 8-Cineole (eucalyptol), a mosquito feeding and 
ovipositional repellent from volatile oil of Hemizoniafitchii (Asteraceae), J. Chem. Ecol., 13,2131,1987. 


2006 by Taylor & Francis Group, LLC 



144 


Insect Repellents: Principles, Methods, and Uses 


41. V. V. Kelkar, A. P. Pandya, andN. R. Metha, Biting activity by Aedes aegypti mosquitoes in guinea pigs: 
an experimental method for screening the effect of some systemically administered compounds, Trop. 
Geogr.Med., 31,415, 1979. 

42. J. C. P. Dias and J. C. daSilva, Some aspects of defensive prophylaxis in Chagas' disease [in Portuguese], 
Rev. Inst. Med. Trop. Sao Paulo, 11, 236, 1969. 

43. A. K. Tripathi et al., Piperitenone oxide as toxic, repellent, and reproduction retardant toward malarial 
vector Anopheles stephensi (Diptera: Anophelinae), J. Med. Entomol., 41,691, 2004. 

44. M. Bar-Zeev and C. N. Smith, Action of repellents on mosquitoes feeding through treated membranes or 
on treated blood, J. Econ. Entomol., 52,263, 1959. 

45. Kingscote, A. A., Orally administered insect repellents: approaches and problems related to the search, 
Proc.IUlnt. Congr. Entomol, 199, 1956. 

46. M. Bar-Zeev and S. Sternberg, Testing space repellents against mosquitoes, Mosq. News, 30, 27, 1970. 

47. R. A. Wirtz et al.. Laboratory testing of repellents against the sand fly Phlebotomus papatasi (Diptera: 
Psychodidae), J. Med. Entomol., 23, 64, 1986. 

48. R. Wasicky, O. Unti, and A. da Silva Ramos, Repellents for mosquitoes [in Portuguese], AnaisFac. Farm, 
e Odont. Univ. Sao Paulo, 7,297, 1949. 

49. S. K. Sharma, V. K. Dua, and V. P. Sharma, Field studies on the mosquito repellent action of neem oil, S. E. 
Asian J. Trop. Med. Public Health, 26, 180, 1995. 

50. L. L. Robert, I. Schneider, and R. A. Wirtz, Deet and permethrin as protectants against malaria-infected 
and uninfected Anopheles stephensi mosquitoes, J. Am. Mosq. Control Assoc., 7, 304, 1991. 

51. Y. Okada et al.. Amino acids and peptides, XI. Synthesis of attractant and repellent peptides for Aedes 
aegypti and Blatella germanica, Chem. Pharm. Bull., 32,4608, 1984. 

52. W. Neng et al., Laboratory evaluation of efficacy of bednets impregnated with pyrethroids, J. Am. Mosq. 
Control Assoc., 7, 294, 1991. 

53. E. B. Starnes and P. Granett, A laboratory method for testing repellents against biting flies, J. Econ. 
Entomol., 46,420, 1953. 

54. M. J. R. Healy, A method for comparing fly-repellant sprays. Biometrics, 9, 290, 1953. 

55. P. W. Wood, The effect of ambient humidity on the repellency of ethylhexanediol (“6-12”) to Aedes 
aegypti. Can. Entomol, 100, 1331, 1968. 

56. M. Kalyanasundarum, A preliminary report on the synthesis and testing of mosquito repellents, Indian 
J. Med. Res., 76,190,1982. 

57. C. F. Curtis et al.. The relative efficacy of repellents against mosquito vectors of disease, Med. Vet. 
Entomol, 1, 109, 1987. 

58. P. Martin and P. B ateson. Measuring Behaviour, an Introductory Guide, 2nd ed., Cambridge: Cambridge 
University Press, 1993. 

59. V. G. Dethier, Notes on the biting response of tsetse flies. Am. J. Trop. Med. Hyg., 3, 160, 1954. 

60. R. Galun, E. M. Kosower, and N. S. Kosower, Effect of methyl phenyldiazencarboxylate (azoester) on the 
feeding behaviour of blood sucking invertebrates. Nature, 224, 181, 1969. 

61. D. J. Fryauff et al.. Contact toxicity of permethrin-impregnated fabric to Hyalomma anatolicum 
excavation (Acari: Ixodidae): Effects of laundering and exposure and recovery times, J. Med. 
Entomol., 35, 335,1998. 

62. M. Bar-Zeev and S. Gothilf, Laboratory evaluation of tick repellents, J. Med. Entomol, 10,71, 1973. 

63. M. Bar-Zeev and S. Gothilf, Laboratory evaluation of flea repellents, J. Med. Entomol., 9, 215, 1972. 

64. A. A. Khan, H. I.Maibach, andD.L. Skidmore, Insect repellents: Effect of mosquito and repellent-related 
factors on protection time, J. Econ. Entomol. 68,43, 1975. 

65. D. R. Barnard et al., Mosquito density, biting rate and cage size effects on repellent tests, Med. Vet. 
Entomol., 12,39, 1998. 

66. L. C. Rutledge et al.. Efficacy of diethyl methylbenzamide (deet) against Aedes dorsalis and a comparison 
of two end points for protection time, J. Am. Mosq. Control Assoc., 5, 363,1989. 

67. H. Lai, S. Ginocchio, and E. J. Hawrylewicz, Procedure for bioassaying mosquito repellents in laboratory 
animals, Proc. Soc. Exp. Biol. Med., 113,770, 1963. 

68. P. Kashin and M. L. Kardatzke, An automated repellency assay system, I. Statistical methods, J. Econ. 
Entomol, 62,10, 1969. 


© 2006 by Taylor & Francis Group, LLC 



Animal Models for Research and Development of Insect Repellents for Human Use 


145 


69. P. Kashin and M. L. Kardatzke, Diurnal rhythm, age, and other variables affecting yellowfever mosquito 
avidity and the laboratory assay of repellents, J. Econ. Entomol., 68, 766, 1975. 

70. P. Kashin and B. E. Ameson, An automated repellency assay system, II. A new electronic “bitometer- 
timer”, J. Econ. Entomol., 62, 200, 1969. 

71. N. L. Sachdeva et al.. Laboratory evaluation of some repellents against the chiggers of trombiculid mites, 
Med. J. Armed Forces India , 39, 39, 1983. 

72. O.M. Abu-shady et al.. Some compounds as repellents against insescts, J. Egypt. Soc. Parasitol., 13,223, 
1983. 

73. R. A. Wirtz et al.. Laboratory testing of repellents against the tsetse Glossina morsitans (Diptera: 
Glossinidae), J. Med. Entomol, 22,271, 1985. 

74. W. S. Choi et al.. Repellent activities of essential oils and monoterpenes against Culex pipiens pollens, 
J. Am. Mosq. Control Assoc., 18, 348,2002. 

75. M. Bar-Zeev andD. Ben-Tamar, Evaluation of mosquito repellents, Mosq. News, 31, 56, 1971. 

76. L. C. Rutledge, Some corrections to the record on insect repellents and attractants, J. Am. Mosq. Control 
Assoc., 4,414, 1988. 

77. E. B. Wilson, An Introduction to Scientific Research, New York: Dover Publications, 1990. 

78. Z. Li et al., Studies on the repellent quwenling [in Chinese], Malaria Res., 6, 1974. 

79. A. C. Cohen, Truncated and Censored Samples: Theory and Application. New York: Marcel Dekker, 
1991. 

80. J. D. Kalbfleisch and R. L. Prentice, The Statistical Analysis of Failure Time Data, 2nd ed., Hoboken: 
Wiley, 2002. 

81. L. C. Rutledge et al., Mathematical models of the effectiveness and persistence of mosquito repellents, 
J. Am. Mosq. Control Assoc., 1, 56, 1985. 

82. T. Poston and I. Stewart, Catastrophe Theory and its Applications, Mineola, NY: Dover Publications, 
1996. 

83. Y. Shirai et al., L-Lactic acid as a mosquito (Diptera: Culicidae) repellent on human and mouse skin, 
J. Med. Entomol., 38,51, 2001. 

84. R. G. D. Steel and J. H. Torrie, Principles and Procedures ofStatistics: A Biometrical Approach, 2nded., 
New York: McGraw-Hill, 1980. 

85. H.C. Fryer etal.. Techniques for conducting fly-repellency tests on cattle, / Econ. Entomol. ,36,33,1943. 

86. H. C. Fryer, F. W. Atkeson, andR. C. Smith, Comparison of methods for testing repellent type fly sprays, 
J. Econ. Entomol., 41, 80, 1948. 

87. F. M. Wadley, Incomplete-block design adapted to paired tests of mosquito repellents. Biometrics Bull., 2, 
30, 1946. 

88. R. M. Altman and C. N. Smith, Investigations of repellents for protection against mosquitoes in Alaska, 
1953, J. Econ. Entomol., 48, 67, 1955. 

89. American Society for Testing and Materials, Standard Method of Field Testing Topical Applications of 
Compounds as Repellents for Medically Important and Pest Arthropods (Including Insects, Ticks, and 
Mites), Philadelphia: American Society for Testing and Materials, 1983. 

90. G. H. Y eoman et al., Repellency of the butyl ester of 3-methylcinchonic acid against Stomoxys calcitrans, 
the stable fly. Vet. Rec., 83, 131, 1968. 

91. L. L. Robert et al.. Comparative sensitivity of four Anopheles (Diptera: Culicidae) to five repellents, 
J. Med. Entomol., 28,417, 1991. 

92. D. J. Finney, Probit Analysis, 3rd ed., London: Cambridge University Press, 1971. 

93. A. Goldstein, Biostatistics: An Introductory Text, New York: Macmillan, 1964. 

94. A. F. Naylor, Comparisons of regression constants fitted by maximum likelihood to four common 
transformations of binomial data, Ann. Hum. Genet., 27, 241, 1964. 

95. American Society for Testing and Materials, Standard Test Methods for Laboratory Testing of 
Noncommercial Mosquito Repellant Formulations on the Skin, Philadelphia: American Society for 
Testing and Materials, 1983. 

96. L. C. Rutledge et al., A field trial of ethyl hexanediol against Aedes dorsalis in Sonoma County, California, 
J. Am. Mosq. Control Assoc., 5, 374, 1989. 

97. W. E. Deming, Statistical Adjustment of Data, New York: Dover Publications, 1964. 


2006 by Taylor & Francis Group, LLC 



146 


Insect Repellents: Principles , Methods, and Uses 


98. L. C. Rutledge et al.. Evaluation of the laboratory mouse model for screening topical mosquito repellents, 
J. Am. Mosq. Control Assoc., 10,565, 1994. 

99. Von R. Wiesmann and R. Lotmar, Observations and investigations on the effectiveness of the new “Kik- 
Geigy” repellent [in German], Acta Trop., 6,292, 1949. 

100. P. I. Rakhmanova, A method of initial selection of mosquito repellents using animals under laboratory 
conditions [in Russian], Med. Parazitol. Parazitar. Bolezni., 28,620, 1959. 

101. O. V. Smirnov, V. S. Suvorov, and H. P. Bocharov, New tests of certain repellents [in Russian], Med. 
Parazitol. Parazitar. Bolezni, 30, 613, 1961. 

102. N. V. Troitskii, Experimental use of carbon disulfide, chloropicrin, K-soap and other insecticides to 
destroy and repel Ornithodorospapillipes [in Russian], Med. Parazitol. Parazitar. Bolezni, 14,75,1945. 

103. M. Y. Liu, Mosquito repellent activity of a series of 2-(lV-alkylamino) acetaldehyde dialkyl acetals and 
their A-acyl derivatives [in Chinese], Sinozoologica, 1,205,1981. 

104. J. Mandel, The Statistical Analysis of Experimental Data, New York: Dover Publications, 1964. 


© 2006 by Taylor & Francis Group, LLC 



$ 


Techniques for Evaluating Repellents 


John M. Govere and David N. Durrheim 


CONTENTS 

Introduction.147 

Important Variables When Evaluating Repellents.149 

Evaluating Repellents Against Blood-Feeding Arthropods.151 

Ethical Considerations.151 

Efficacy Testing.152 

Amount of Repellent.153 

Data Recording.153 

Evaluating Compounds as Repellents of Mosquitoes and Biting Flies.153 

Evaluating Compounds as Repellents of Crawling Arthropods.154 

Fleas.154 

Ticks.155 

Chigger Mites.156 

General Considerations for Evaluating Repellent Candles, Coils, and Vaporizing Mats.156 

General Considerations for Evaluating Treated Articles or Clothing.156 

General Principles for Obtaining Valid and Reliable Results.157 

References.157 


Editors’ note : Drs. Govere and Durheim have produced a comprehensive guide to repellent product 
testing based on their own experiences in the field, including a justification of the process based on a 
partial review of medical entomology. The chapter is presented in the form of definite steps for a 
successful test based on specific assumptions. Some of the procedures recommended by the authors are 
controversial (e.g., the use of a subject as his own control or rinsing the skin with alcohol prior to 
application of the repellent product). The editors realize that, in reality, the assumptions are often 
violated and that every technique is a compromise between practicality and precision. Nonetheless, the 
chapter should be valuable as a foundation for considerations necessary for reliable evaluations of 
repellent products. 


Introduction 

Arthropod-transmitted pathogens remain a major source of morbidity and mortality worldwide. 1 
The wide array of arthropod-borne pathogens constitute an enormous public health burden, particularly 


147 


© 2006 by Taylor & Francis Group, LLC 























148 


Insect Repellents: Principles , Methods, and Uses 


TABLE 8.1 


Medically Important Biting Diptera and Other Arthropods 


Vectors 

Diseases 

Mosquitoes (Culicidae) 

Anopheles 

Malaria, lymphatic filariasis 

Culex 

Lymphatic filariasis, Japanese encephalitis, certain arboviruses 

Aedes 

Yellow fever, Lympatic filariasis, dengue fever, certain arboviruses 

Mansonia 

Lymphatic filariasis 

Other biting Diptera 

Tsetse flies ( Glossina) 

African sleeping sickness 

Blackflies ( Simulium) 

River blindness (onchocerciasis) 

Sandflies (Phlebotomus, Lutzomyia ) 

Leishmaniasis, sandfly fever 

Horseflies ( Tabanidae) 

Loiasis, tularaemia 

Stable flies ( Stomoxys) 

Biting midges ( Ceratopogonidae ) 

Mansonellosis 

Other biting Arthropods 

Fleas ( Siphonaptera) 

Plague, flea-borne typhus 

Chigoe Fleas (Tunga penetrans) 

Jigger infection 

Head louse ( Pediculus humanus capitis) 

Pediculosis 

Body louse ( Pediculus humanus humanus) 

Trench fever, louse-borne relapsing fever, epidemic typhus 

Pubic louse (Pthirus pubis) 

Hard ticks ( Ixodidae ) 

Arboviral encephalitis and fevers, Lyme disease 

Soft ticks (Argasidae) 

Tick-borne relapsing fever 

Biting mites (Trombiculidae) 

Scrub typhus 


in developing tropical countries (Table 8.1). Mosquitoes transmit pathogens to more than 700 million 
people annually and the diseases they cause are estimated to be responsible for one out of every 17 deaths 
globally. 2 Malaria is undoubtedly the most important of the diseases caused by mosquito-transmitted 
pathogens; it is responsible for as many as three million deaths and 500 million episodes of illness each 
year. 3 Mosquitoes also transmit the arboviruses responsible for yellow fever, dengue hemorrhagic fever, 
epidemic polyarthritis (including Ross River and Barmah Forest viruses), and several forms of 
encephalitis. The filarial nematodes are another group of pathogens transmitted by mosquitoes. These 
parasites cause lymphatic filariasis, the second most common cause of chronic disability world-wide. 4 

Ticks are vectors of a large number of diseases of animal and humans. Human diseases associated with 
pathogens transmitted by ticks include tick-borne relapsing fever, Rocky Mountain spotted fever, Q fever, 
Lyme disease, and others. Chiggers are both the reservoir and the vector for the pathogen that causes scrub 
typhus, a disease that accounts for up to 20% of all fever presentations in parts of Asia. 5 Fleas are also 
important vectors of disease, including bubonic plague and flea-borne endemic typhus. 4 In addition to 
disease, many insects, including mosquitoes, biting flies, fleas, and ticks, are a source of mental 
anguish—causing intense annoyance and sleep disturbance. Although most biting Diptera also feed on 
plant juices, females generally need a blood meal for egg development. 6 

Given the enormous burden of disease resulting from arthropods it is not surprising that humankind 
has invested in a vast assortment of methods for controlling insect pests and vectors. Personal protection 
has become an increasingly popular method for preventing contact with arthropods, as community vector 
control is not always available. There is an understanding that even if personal protection is not 
completely effective in preventing exposure to infectious biting arthropods, it can significantly reduce 
personal risk by decreasing the chance of being bitten by an infectious vector, complementing other 
strategies to reduce risk. 

Unfortunately, many commercially available products, as well as traditional home remedies that are 
more affordable in some regions, are not very effective. As a false sense of security based on application 
of an ineffective repellent may have devastating consequences, it is crucial that all products alleged to 


© 2006 by Taylor & Francis Group, LLC 





Techniques for Evaluating Repellents 


149 


have repellent properties are rigorously evaluated, preferably in a standardized fashion that facilitates 
comparisons between studies. 

An ideal insect repellent should be active against multiple species of biting arthropods, remain 
effective for at least eight hours, cause no topical irritation of the skin or mucous membranes, cause no 
systemic toxicity, resist physical removal by rubbing, and be greaseless and odorless. If a repellant is to 
be used on surfaces or materials, it should not be abrasive or damaging to the target material. This chapter 
introduces approaches that have proven valuable in evaluating repellents against a number of arthropod 
pests and vectors. 


Important Variables When Evaluating Repellents 

The attractiveness of different persons to insects varies substantially. 7-9 Adults are more likely to be 
bitten than children. 10 ' 11 Men are more readily bitten than women and larger persons attract more insects, 
perhaps because of greater relative heat and carbon dioxide output. 1 " 13 Recent research suggests that 
malaria-infected children with gametocytes are more attractive to malaria vectors. 14 

Biting insects are attracted to dark clothing, carbon dioxide, lactic acid, floral or fruity fragrances, skin 
temperature, and moisture. 15-17 Mosquitoes become restless when there is an increase of carbon dioxide 
in their vicinity; they tend to fly in the direction of the carbon dioxide. Sensory hairs on the mosquito’s 
antennae detect changes in carbon dioxide content, humidity and temperature of the air. If the sensors 
detect a decrease in carbon dioxide, humidity, or temperature, the mosquito turns aside. Repellent 
molecules block receptor sites in mosquito sensory hairs causing a mosquito to avert their potential 
human target. 18 

The environment, and particularly climatic conditions, can affect biting behavior. This is a significant 
issue during field trials when wind, cooler temperature, and rainfall can markedly decrease feeding. 
Seasonal fluctuation in vector abundance and population fluctuations in response to environmental 
factors are also important considerations in attempting to conduct and interpret field trials. 

Additionally, an understanding of vector feeding preferences is notable. This includes preferences 
for feeding indoors or outdoors as well as preferences for feeding at certain times of day or night 
(Table 8.2). 

Insect bites are not randomly distributed on the human body (Table 8.3). This concept has been most 
thoroughly studied in relation to mosquito disease vectors. Anopheles gambiae s.s. prefers to bite the feet 
of seated humans and Anopheles arabiensis mosquitoes prefer to bite the ankles and feet of motionless 
humans. 919-21 The absolute number of Anopheles arabiensis mosquitoes biting motionless humans 
drops dramatically when their feet are covered with shoes without a significant shift to other parts of the 
leg or remainder of the body. 22 Simulium damnosum in West Africa predominantly bites on the leg. 2 ' 3 
Aedes simpsoni prefers to bite the face of naked humans, while Eretmopodites chrysogaster prefers to 
bite ankles and feet. 24 Sabethes belisarioi appears to exclusively bite human noses. 25 Anopheles 
albimanus mosquitoes prefer to bite the head and neck. Anopheles atroparvus prefers to bite the head 
and shoulders, and Culex pipiens and Culex quinquefasciatus bite on the lower half of the body. 26-28 
Anopheles farauti prefers to bite near the ground and Anopheles gambiae s.s has a strong preference for 
the feet of seated humans. 29 Anopheles arabiensis has a strong biting preference for the ankles and feet of 
individuals sitting on camp chairs. 9 ’ 30,31 

An understanding of the various preferences involved in mosquito feeding behavior should be woven 
into field trials, as these characteristics can provide opportunities for targeted prevention efforts. For 
example, in a South African randomized cross-over study of Anopheles arabiensis it was demonstrated 
that mosquito bites could be reduced by almost 70% when only the ankles and feet were treated with deet 
repellent (Table 8.4a and Table 8.4b). 9 The application of affordable repellents could certainly 
complement the use of other techniques for control of a number of vector-borne pathogens, particularly 
those in which human behavior and mosquito feeding temporally intersect in an outdoor environment. 


© 2006 by Taylor & Francis Group, LLC 




150 


Insect Repellents: Principles , Methods, and Uses 


TABLE 8.2 


Typical Biting Behavior, Day/Night and Indoor/Outdoor Preference of Biting Diptera and Other Arthropods 


Pest/V ector 


Blood-Feeding Stage 

Indoor/Outdoor Biting 

Day/Night Biting 

Anopheles 

Female 

adult 



Indoor 

Night 

Culex 

Female 

adult 



Indoor 

Night 

Aedes 

Female 

adult 



Outdoor/indoor 

Day 

Mansonia 

Female 

adult 



Indoor 

Night 

Tsetse flies 

Female 

and male 

adults 


Outdoor 

Day 

Triatomine bugs 

Female 

and male 

nymphs 

and adults 

Indoor/outdoor 

Night 

Bedbugs 

Female 

and male 

nymphs 

and adults 

Indoor 

Night 

Black flies 

Female 

adult 



Outdoor 

Day 

Sand flies 

Female 

adult 



Outdoor/indoor 

Night 

Horse flies 

Female 

adult 



Outdoor 

Day 

Stable flies 

Female 

and male 

adults 


Outdoor/indoor 

Day 

Fleas 

Female 

and male 

nymphs 

and adults 

Indoor/outdoor 

Day or night 

Chigoe fleas 

Female 

and male 

nymphs 

and adults 

Indoor/outdoor 

Day or night 

Head louse 

Female 

and male 

nymphs 

and adults 

Indoor/outdoor 

Day or night 

Body louse 

Female 

and male 

nymphs 

and adults 

Indoor/outdoor 

Day or night 

Pubic louse 

Female 

and male 

nymphs 

and adults 

Indoor/outdoor 

Day or night 

Hard ticks 

Female 

and male 

nymphs 

and adults 

Indoor/outdoor 

Day or night 

Soft ticks 

Female 

and male 

nymphs 

and adults 

Indoor/outdoor 

Day or night 

Biting mites 

Female 

and male 

nymphs 

and adults 

Indoor/outdoor 

Day or night 


This knowledge has also been successfully applied, at least on one occasion, to control a focal malaria 
epidemic by topical application of 15% deet to the feet and ankles of an affected community. 32 

The variety of factors that influence the field effectiveness of a repellent—including the frequency and 
uniformity of application, number and species of insects attempting to bite, an individual’s attractiveness 
to blood-sucking arthropods, an individual’s level of physical activity, amount of abrasion of treated skin 
by clothing, evaporation and absorption from the skin surface, wash-off from rain or sweat, prevailing 
temperature, and degree of wind disturbance—must be controlled during evaluation to allow comparison. 
Certain factors are easier to control, like abrasion, which can be reduced by limiting movement of the 
experimental subjects. Others, including interpersonal differences, may be controlled by using the 
individual as their own control or through randomized cross-over study designs. Climate variables 
should be carefully recorded and described, and repeated measures made under different conditions to 
ensure that findings are reproducible. 

TABLE 8.3 


Anatomical Biting Preferences of Selected Mosquitoes on Humans 


Mosquito Species 

Site 

Reference 

Eretmopodites chrysogaster 

Ankles & feet 

24 

Aedes simpsoni 

Face 

24 

Sabethes belisarioi 

Human noses 

25 

Anopheles albimanus 

Head & neck 

26 

Anopheles atroparvus 

Head & shoulders 

23 

Aedes aegypti 

Head & shoulders 

23 

Culex pipiens 

Lower half of body 

28 

Culex quinquefasciatus 

Lower half of body 

28 

Anopheles farauti 

Near ground level 

29 

Anopheles gambiae s.s. 

Near ground level 

20 

Anopheles arabiensis 

Ankles & feet 

31 

Anopheles arabiensis 

Ankles & feet 

32 


© 2006 by Taylor & Francis Group, LLC 




Techniques for Evaluating Repellents 


151 


TABLE 8.4a 


Distribution of Mosquito Bites on the Human Body when Ankles and Feet Were Untreated 


Body Part 



Nights 



i 

2 

3 

4 

5 

Total (%) 

Ankle/feet 

48 

54 

145 

102 

72 

421 (81.1) 

Legs 

5 

8 

30 

31 

11 

85 (16.4) 

Arms 

1 

0 

1 

4 

1 

7(1.3) 

Body 

1 

1 

3 

1 

0 

6 (1.2) 

Total (%) 

55 (10.6) 

63 (12.1) 

179 (34.5) 

138 (26.6) 

84 (16.2) 

519 (100.0) 

Source : J. 

M. Govere, L. E. O 

'. Braack, D. N. Durrheim, R. H. Hunt, and M. Coetzee Med. Vet. Entomol.. 15, 287, 2001. 


TABLE 8.4b 

Distribution of Mosquito Bites on the Human Body when Ankles and Feet Were Treated with Deet 

Body Part 



Nights 



1 

2 

3 

4 

5 

Total (%) 

Ankle/feet 

0 

0 

0 

0 

0 

0 (0.0) 

Legs 

32 

15 

28 

39 

27 

141 (88.1) 

Arms 

0 

0 

2 

2 

0 

4 (2.5) 

Body 

3 

0 

9 

1 

2 

15 (9.4) 

Total (%) 

35 (21.8) 

15 (9.4) 

39 (24.4) 

42 (26.3) 

29 (18.1) 

160 (100.0) 

% Protection 

36.4 

76.2 

78.2 

69.6 

65.5 

69.2 


Source: J. M. Govere, L. E. O. Braack, D. N. Durrheim, R. H. Hunt, and M. Coetzee Med. Vet. Entomol ., 15, 287, 2001. 


Evaluating Repellents Against Blood-Feeding Arthropods 

Repellent testing procedures are conducted through both laboratory evaluation and field testing; 
although a myriad of testing procedures have been described, few have been widely used. Most 
validated procedures relate to medically important arthropods with a special emphasis on mosquito 
repellent testing. This chapter will focus on generally accepted, reliable and practical approaches 
for testing the performance of compounds that purport to repel mosquitoes, biting flies, fleas, 
chiggers, and ticks from human skin or from the environment near people. As a variety of 
formulations are possible, including liquid or pressurized products for spray treatments, material or 
article impregnation, lotions, coils, candles or vaporizing mats, testing must include the end-use 
product formulation. 


Ethical Considerations 

It is generally accepted as unethical to expose a person to an experimental chemical compound without 
fully informed consent from that subject. Information provided to the volunteer subject must include a 
full description of the nature and purpose of the test, and any physical or mental health consequences that 
are reasonably foreseeable. The subject must be guaranteed the option to withdraw at any stage without 
prejudice. It is essential to ensure that the insects used are not infected with known human pathogens. 
Some of the arguments against using humans as test subjects in laboratory and field tests include 


2006 by Taylor & Francis Group, LLC 



152 


Insect Repellents: Principles , Methods, and Uses 


concerns regarding ethical considerations, sensitization reactions following tick, flea or mite bites, and 
poorly established toxicity profiles of chemicals. It is therefore important that the toxicokinetics 
(absorption, biodistribution, metabolism, and excretion) of test compounds be determined in animal 
models. However, products developed for human use should, whenever possible in the course of 
development and evaluation, be tested on people. 

Although it may be considered preferable to fully assess the safety profile of a product, including 
potential for topical irritation, prior to exploring its repellency, initial laboratory screening of promising 
candidates on a human arm is often conducted without prior skin irritation studies. Treated cotton 
stocking is worn over the arm or a treated cloth is tested over untreated cloth that covers the skin surface 
so that chemicals are not in direct contact with the skin surface. 33,34 


Efficacy Testing 

The number of test subjects required depends on the purported duration of effect. For a product with 
1-4 h of repellency, at least five treated test subjects should be used. For a label claim of 5 or more 
hours of repellency, at least 10 treated test subjects should be used. Similar numbers of adult male 
and female test subjects are preferable. Test subjects should not ingest alcohol or caffeine and avoid 
applying fragrant products (e.g., perfumes, colognes, hair sprays, and lotions) for at least 12 h before 
the testing. 

The behavior of the species of biting arthropods that are the subject of the study should be examined 
during the trial. Biting frequency on untreated skin is used to determine avidity of flying insects. Tick 
drags made of white flannel cloth can be pulled over the ground and low vegetation to identify heavy tick 
infestations. Chigger mites are located by laying black plates on the ground. 

Ideally, untreated subjects should be used as controls. Test subjects should be at least 3 m apart during 
the test and may engage in usual outdoor activity, including non-vigorous movements like intermittent 
slow walking, standing, squatting, sitting, and raising or lowering arms. Tobacco should not be used 
during testing. 

Many studies have been based on treatment of the test subject’s forearm (wrist to elbow), but the 
lower leg may also be used. The exposed surface area (in cm 2 ) of each test subject should be carefully 
calculated, by measuring the circumference of the arm at the wrist, the elbow, and three to four equally 
spaced points in-between, and then multiplying the average of these circumference measures by the 
distance from the wrist to elbow; the same method for calculating surface area can also be applied to 
the lower leg, measuring from the ankle to the knee. The upper arm or leg and hand or foot should be 
covered with a material impenetrable to the insect’s proboscis. Dark colors should be avoided and latex 
gloves may be used to cover the subject’s hands. The test area should be washed with unscented soap, 
rinsed first with water and then with a solution of 70% ethanol in water, and finally dried with 
a clean towel. 

A test subject should receive no more that one treatment per test, potentially replicated on each limb. 
Test subjects should avoid exertion, which might increase perspiration or abrasion. The treated area 
should also not be rubbed, touched, or wetted. Other body parts, including the face, back, and non-test 
limbs should be adequately protected with gloves, head net, and protective clothing so that biting 
pressure is concentrated on the exposed treated skin. Field evaluation of repellents on skin should only be 
conducted after favorable toxicology has been established for the test chemicals. 

A subject’s forearms can be used in paired tests to determine protection time, which is calculated 
to the first bite, and confirmed by second and third bites within 5 min. A number of factors can affect 
results, including the species being evaluated, the density of insects, age and gonotrophic state of the 
insect population, age of host, time of day, and temperature and humidity. To effectively deal with 
multiple factors in the analysis, a larger number of test subjects and more test replications 
are necessary. 


© 2006 by Taylor & Francis Group, LLC 



Techniques for Evaluating Repellents 


153 


Amount of Repellent 

The test formulation should be stored at room temperature and ambient humidity before testing. The 
time since production of the test formulation should be recorded. Generally, products should be less 
than 12 months old. Standard application quantities are 1 g of liquid aerosol or pump spray test 
material, or 1-1.5 g of cream, lotion or stick, per 600 cm 2 of test area. This should be evenly applied to 
the forearm or lower leg. It is important to confirm that the quantity per unit area does not vary by more 
than 5% for all subjects or replicates. 

Data Recording 

The number of bites and probes should be recorded by the investigator rather than the test subject. The 
duration of repellent protection should be recorded for each test subject and for each test site. 
Traditionally test results are reported for complete protection time and 95% repellency. 

Complete protection time is a measure of the duration of repellent protection until the time of first bite 
for each test subject, usually confirmed by a second bite occurring within the same time interval. The 
mean protection time and standard error are then calculated for each test species across subjects. 
Statistical testing should be used to examine variability between repetitions and between means. 

The duration of repellent protection based on the period with 95% reduction in bites for each 
test subject is referred to as 95% repellency. The mean protection time and standard error based on a 
95% reduction in bites for each test species should be reported. Statistical testing may be used for 
examining variability between repetitions and means. Survival analysis may also be used when 
comparing multiple products. It is important that the choice of statistical method selected be 
clearly explained. 


Evaluating Compounds as Repellents of Mosquitoes and Biting Flies 

There is a rich tradition of testing mosquito repellents, with the first well-planned laboratory evaluation 
conducted in 1919. 35 The principal mosquito specie used in tests for mosquito repellency is Aedes 
aegypti, which is relatively easily reared and maintained, and an avid blood feeder even in the laboratory. 
However, compounds have differential effectiveness against other vector species. For example, 
a repellent considered poorly effective against Aedes aegypti was found to be highly effective against 
deer flies ( Chrysops spp.) 36 It is therefore important to test compounds against the specific target species 
of interest. 

Generally, laboratory testing of mosquitoes should include at least three genera of human biters; 
Aedes aegypti, an Anopheles species, and a Culex species. When reporting on either laboratory or 
field testing, it is important to identify test insects by genus and species, and by subspecies or strain, 
particularly with mosquitoes. With field testing, identity of the insect should be confirmed by 
aspirating specimens into a vial for laboratory identification before testing begins. Biting pressure 
should be periodically determined throughout the test. Laboratory mosquitoes should be adult females 
5-10 days old. Stable flies should be 3 days old. The age or age range of the test insects should 
be reported. 

Larvae should be reared in the laboratory under optimal conditions for the particular species. As 
a general guide, most species should be reared at 27 + 3°C, with a relative humidity of 80+ 10%, and a 
photoperiod of 16:8 h (light:dark). Other conditions may be used where appropriate for a particular 
species, with any alternative rearing techniques justified in the study summary. Adults should be fed 10% 
sucrose and no blood meal should be offered before the test. Test insects should be starved for 12 h 
immediately before the test, used for only a single test, and destroyed after the trial. 


2006 by Taylor & Francis Group, LLC 



154 


Insect Repellents: Principles , Methods, and Uses 


It is common practice that at least one mosquito should be introduced for every 100 cm 3 of cage space 
with at least 200 mosquitoes introduced in each test cage. Similarly, for stable flies, at least one stable fly 
is commonly introduced for each 500 cm 3 of cage space with at least 45 stable flies introduced in each 
test cage. 

Test cages should be at least 20,000 cm in volume, square or rectangular in shape, and feature one 
sleeved opening for the subject’s arm. Each cage should be used for only one test subject and treatment at 
a time. Temperature should be maintained at 22-27°C, relative humidity at 50% to 80%, and lights 
should be kept on during testing. 

An untreated (negative) control is recommended to verify biting pressure. There is some debate as to 
the most appropriate control, with protagonists arguing in favor of either using the untreated forearm or 
lower leg of the same test subject, or another untreated individual. Both approaches have advantages and 
disadvantages. The control forearm or lower leg should be prepared, washed, rinsed, and dried in 
precisely the same manner as the treated forearm or lower leg. Before the test begins, subjects should 
expose their untreated forearm to the mosquitoes or stable flies in the test cage to establish their 
attractiveness. It is recommended that a minimum of 10 mosquitoes land and probe within 30 s, or five 
stable flies land and probe within 60 s, for a subject to participate. Every hour, an untreated forearm or 
lower limb should be inserted through the sleeve into the cage and exposed to mosquitoes for up to 30 s, 
or to stable flies up to 60 s, to verify biting pressure. The forearm or lower limb should be removed from 
the test cage as soon as it has received the necessary number of probes. Probing is preferred to biting so 
that a subject’s discomfort is limited. 

Thirty minutes after treatment with repellent, the treated area should be inserted through the sleeve 
into the cage for 5 min. This allows sufficient time for the repellent to dry and still tests the minimum 
reasonable protection time that might be of practical value. The number of bites or probes in each 
exposure period should be recorded. The treated area should be exposed for 5 min every 30 min while 
biting pressure lasts, i.e., until the control area no longer receives 10 mosquito landings in 30 s or five 
stable flies landings in 60 s. Test subjects should avoid rubbing their arm or leg when introducing or 
removing it from the cage and between exposure periods. 

An alternative approach, particularly when comparing products, is to insert the treated limb into the 
cage for one minute, and if not bitten, to reinsert the limb for one minute every 5 min, up to 2 h. If biting 
still does not occur, then the interval can be extended to 15 min. If at any point mosquitoes begin landing 
but not biting (a behavior that occurs when the efficacy of a repellent begins to wane), then the intervals 
between insertions could be reduced to 5 min. 

Field testing should be conducted at a minimum of two field sites in environmentally distinctive 
habitats (e.g., forest, grassland, salt marsh, wet land, beach, barns, or an urban environment) suitable for 
the target insect. For mosquitoes, different species prefer various habitats. Habitats where biting pressure 
is below the levels described previously are unlikely to provide reliable and reproducible results. It is 
important to record details of weather conditions during the test, including temperature, relative 
humidity, cloud cover, precipitation, light intensity, and wind speed, allowing 90 s of observation for 
each exposure period. It is important that wind speed does not exceed 10 mph as windy conditions cause 
diminished probing. 


Evaluating Compounds as Repellents of Crawling Arthropods 
Fleas 

The cat flea, Ctenocephalides felis, is the preferred model flea for repellent testing. Adult male and/or 
female fleas that are 5-10 days old, reared at 27 + 3°C, with a relative humidity of 80+10%, and a 
photoperiod of 16:8 h (light:dark) should be used. The adult fleas should not be blood-fed, and after one 
trial they should be destroyed. There should be one flea per 9 cm 3 and at least 100 fleas in each test cage. 


© 2006 by Taylor & Francis Group, LLC 



Techniques for Evaluating Repellents 


155 


Twenty-five fleas should be added to the test cage after each exposure period. Cages should be at least 
900 cm 3 in volume, square, circular, or rectangular in shape, and made of plastic or glass with an opening 
on top to insert the test subject’s limb. The cages should have a rough floor utilizing a material such as 
clean sand. Replications should be limited to one test subject and treatment at a time for each cage. 
The temperature during testing should be maintained at 22-27°C, with relative humidity at 50-80%, 
and the lights should be on. 

A negative (untreated) control is recommended to verify biting pressure, with the negative control 
being either the test subject’s untreated forearm or lower leg, or an untreated subject. The control limb 
should be washed, rinsed, and dried in exactly the same way as the treated area. Before testing begins, 
subjects should expose their forearms to the fleas in the test cage to establish their attractiveness. It is 
recommended that to qualify as a participant, a subject should experience a minimum of 10 flea landings 
or probes within 30 s. Every hour, a control limb should be inserted through the sleeve into the cage and 
exposed to the fleas for up to 30 s to verify biting pressure, with the limb removed as soon as 10 landings 
have occurred. 

Thirty minutes after treatment with repellent, the test subject’s forearm or lower leg should be inserted 
through the sleeve into the cage for 5 min and the number of landings recorded for each exposure period. 
This should be repeated every 30 min while biting pressure lasts, that is, until the control no longer 
receives 10 flea landings in 30 s. The duration of repellent protection for each test subject should be 
recorded and the mean protection time and standard error reported. 

An alternative method for testing compounds against fleas requires use of two strips of fabric—one 
impregnated with a chemical and one untreated control. These are lowered into a container into which 
fleas have been added. 37 After a predetermined time, the strips are removed, the fleas remaining on the 
cloth strips are counted, and the percentage of repellency is calculated. Use of an olfactometer for 
comparative testing of new repellents against fleas has also been described. 38 

Ticks 

Animals, such as gerbils, have been used for evaluating repellents against crawling insects. 39 The 
animal may be immobilized in a stanchion with its shaved abdomen exposed and two identically sized 
areas treated with a candidate repellent and a standard. An alternative strategy is to dip or spray the 
caged and restrained animal and then place it on the periphery of a tick or mite-infested area. 
Following a specific exposure time, the engorged and attached ticks or mites are counted to determine 
protection afforded by the repellent. Results from laboratory testing may not always correlate to field 
evaluation. Similarly, performance in animal testing may not fully correlate to performance 
on humans. 

Tick species used for evaluation should be disease free and represent both ixodid (hard) and argasid 
(soft) ticks. Adult and nymphal ticks should be tested, since both life stages can be involved in pathogen 
transmission. Ticks should be reared at 22 + 3°C, with a relative humidity of 50-80%, and a photoperiod 
of 16:8 h (light:dark). Ticks used for testing should be destroyed after a single trial. Five ticks should be 
exposed to the treated forearm or lower leg in each exposure period, keeping the temperature during the 
test at 22-27°C, with a relative humidity of 50-80%, and the light on. The duration of repellent protection 
for each subject should be recorded. 

As an alternative to direct testing on humans, treated cloth patches may be placed on a paddle and 
touched to the bottom of a pen infested with ticks or mites. The number of ticks or mites crawling from 
the untreated part of the paddle to a point midway up the treated patch allows evaluation of repellency 
when compared with an untreated control paddle. 

A means for determining the minimum effective dosage of a repellent against ticks has also been 
described. 39 The candidate material is applied in horizontal stripes of progressively increasing 
concentrations. Ticks then climb the vertically positioned fabric until they reach a concentration they 
cannot tolerate, as indicated by the ticks dropping off of the surface. 


2006 by Taylor & Francis Group, LLC 



156 


Insect Repellents: Principles , Methods, and Uses 


Chigger Mites 

Chigger mites include a number of genera in the family Trombiculidae, but most tests have involved 
either Eutrombicula or Leptotrombidium. Larval, unfed chigger mites reared at 22 + 3°C, at a relative 
humidity of 50-80%, and a photoperiod of 16:8 h (light:dark) should be used. The age of the mites should 
be recorded. Mites used in a trial should be destroyed afterwards. Five mites should be exposed to the 
treated area of skin during each exposure period, while keeping the temperature at 22-27°C, relative 
humidity at 50-80%, and the light on. 

The area above and below the test area should be covered with material that chigger mite mouthparts 
cannot penetrate. A negative (untreated) control, usually an untreated surface on the same subject, is 
recommended to verify biting pressure. Test organisms should be picked up with a soft artist’s paintbrush 
and placed on the test subject about 2 cm from the area of the forearm where the repellent has been 
applied, near the wrist, with a new tick or mite placed 2 cm below the test area once the previous 
specimen has crossed onto the test area. After moving toward the margin of the test area, chigger mites 
should be allowed 5 min to cross the margin onto the test material (toward the elbow). Once the chigger 
mite has been recorded as not repelled, it should be replaced with one that was not previously tested. 
A new group of chigger mites should be exposed to the test material every 30 min. No test arthropods 
should be reused. 


General Considerations for Evaluating Repellent Candles, Coils, 
and Vaporizing Mats 

The species and biting pressure should be determined, test sites prepared and testing conditions recorded 
as described above. If more than one test subject is exposed to the same candle, coil, or mat, then the 
number of bites should be averaged. The number and placement of the intervention(s) should be 
consistent with label directions or proposed use. Test subjects should be located at the maximum distance 
of usefulness proposed or described. If the product description states that the candle, coil, or mat should 
be placed upwind, then test subjects should remain downwind; otherwise, test subjects should move 
around the circumference of the test area periodically with the time interval of movements reported in the 
study results. 

A negative (untreated) control of the same size as the test area is desirable to determine biting pressure. 
Control subjects should remain upwind, far enough from the treatment area as is necessary so as not to be 
affected by the repellent. They should be exposed for the full period of activity of the candle, coil, or mat. 
For coils, protection time should be the same as burning time. 

An investigator or study partner (not the test subject) should record the number of bites and probes. 
When compared to the negative control, at least 50% of insects should be repelled. The duration of 
repellent protection and mean time to 50% reduction in bites, with standard error, should be reported. 

An alternative approach that has merit for evaluation of area repellent systems is the use of carbon 
dioxide-baited traps. These can be placed in the treated and control areas simultaneously and collections 
made for equivalent time periods to allow for comparison. It is important that catches are correctly 
identified, as the repellent method may have a differential effect against different species. 


General Considerations for Evaluating Treated Articles or Clothing 

Evaluations of repellent impregnated clothing or treated articles should report the repellent used, 
impregnation formulation, method of impregnation, type of garment treated, and amount of repellent 


© 2006 by Taylor & Francis Group, LLC 



Techniques for Evaluating Repellents 


157 


absorbed per unit area of textile. Repellents may be used to treat a variety of materials, including those 
used for making bed nets, tablecloths, loose jackets, and other clothing items. 

Reports of field tests should include details on type of material treated, mesh size for nets, weight 
per unit area of material, impregnation formulation, method of impregnation, amount of repellent 
absorbed per unit area, and method of exposure. The degree of protection between subjects using treated 
articles or clothing should be compared to subjects in the same environment using the same but untreated 
articles or clothing during a standard exposure period. A bite or probe should be recorded whenever an 
arthropod proboscis penetrates the treated material. 


General Principles for Obtaining Valid and Reliable Results 

Evaluations should be as simple and practical as possible to encourage standard comparison and 
universal acceptance by being easily understood and performed. 40 Although candidate repellents should 
be taken to the field to determine protection time and effectiveness under field conditions, it is preferable 
to first conduct evaluations in a laboratory environment as it is easier to control for potential confounding 
factors. 41 Creative design may permit imitation of important individual held conditions in the laboratory. 
A good example is the use of extractor fans to create draught. 42 Testing should be performed on human 
subjects to evaluate actual performance against the target host species. Standardized amounts of test 
compounds should be applied and uniform coverage ensured. 

Where a “gold standard” repellent exists, it is sensible to compare promising candidates to this 
product. It is important that the test subjects and the person recording probes or bites do not know which 
compound has been applied (i.e., a double-blinded trial). Cross-over designs are useful to take account of 
other potential confounding factors, but persistent and longer range effects of repellents must be factored 
into the study design. The sequential application of all repellents to each individual is a preferred 
strategy. 43 It is then important that the sequence of application is randomized and that there is careful 
cleaning of the test area after each application. Ideally, different products should be tested on different 
days on the same individual. Meticulous recording of experimental conditions should allow easy 
replication by other investigators. Biting rates on untreated skin should be recorded to assure adequate 
biting pressure. Three to six replications, preferably on multiple subjects, should be conducted to 
determine interpersonal variability and provide a mean protection time. This variability should also be 
factored into study design and analysis as the differences in interpersonal attraction may be profound. 44 

The potential for repellents to contribute to the integrated control of arthropod borne diseases of 
humans has not yet been fully realized. 45 The availability of affordable alternatives, including mosquito- 
repellent plants or plant-derived natural products may make this complementary strategy more 
feasible. 46 ' 47 However, before widespread use can be encouraged to prevent potentially life-threatening 
diseases, it is essential that the efficacy and duration of effect is determined reliably. 

References 

1. J. R. Roberts and J. R. Reigart, Does anything beat DEET?, Pediat. Ann., 33, 444, 2004. 

2. C. P. McHugh, Arthropods: Vectors of disease agents. Lab. Med., 25, 429, 1994. 

3. World Health Organization, International Travel and Health Vaccination Requirement and Health 
Advice, Geneva: WHO, 1995. 

4. D. N. Diirrheim et al.. Lymphatic filariasis endemicity—an indicator of poverty?, Trop. Med. Int. 
Health, 9, 843, 2004. 

5. D. Strickman, Scrub typhus, in The Encyclopedia of Arthropod-Transmitted Infections of Man and 
Domesticated Animals, M. W. Service (Ed.), Wallingford, U.K., CAB International Publishing, 2001, 
p. 456. 

6. A. N. Clements, The Physiology of Mosquitoes, Oxford: Pergamon Press, 1963. 


2006 by Taylor & Francis Group, LLC 



158 


Insect Repellents: Principles, Methods, and Uses 


7. C. F. Curtis, Personal protection methods against vectors of disease. Rev. Med. Vet. Entomol., 80, 543, 
1992. 

8. A. A. Khan, Mosquito attractants and repellents, in Chemical Control of Insect Behavior, H. H. Shorey 
and J. J. McKelvey (Eds.), New York: Wiley, 1977, p. 305. 

9. J. M. Govere et al.. Effect of repellent on Anopheles arahiensis biting humans in Kruger Park, South 
Africa. Med. Vet. Entomol., 15, 287, 2001. 

10. R. C. Muirhead-Thomson, The distribution of anopheline mosquito bites among different age groups, 
Br. Med. J., 1, 1114, 1951. 

11. I. H. Gilbert, H. K. Gouck, and N. Smith. Attractiveness of men and women to Aedes aegypti and 
relative protection time obtained with DEET, Fla. Entomol., 49, 53, 1966. 

12. G. R. Port and P. F. L. Boreham, The relationship of host size to feeding by mosquitoes of the 
Anopheles gambiae Giles complex (Diptera: Culicidae), Bull. Entomol. Res., 70, 133, 1980. 

13. C. M. Gjullin, Effect of clothing color on the rate of attack of Aedes mosquitoes, J. Econ. Entomol., 40, 
326, 1947. 

14. R. Lacroix et al.. Malaria infection increases attractiveness of humans to mosquitoes, PLoS Biol., 3, 
e298, 2005. 

15. C. E. Schreck, D. L. Kline, and D. A. Carlson, Mosquito attraction to substances from the skin of 
different humans. J. Am. Mosq. Control Assoc., 6, 406, 1990. 

16. R. Brouwer, Variations in human body odour as a cause of individual differences of attraction for 
malaria mosquitoes, Trop. Geogr. Med., 12, 186, 1960. 

17. M. Geier, H. Sass, and J. Boeckh, A search for components in human body odour that attract females 
of Aedes aegypti, in Olfaction in Mosquito-host Interaction, G. R. Bock and G. Cardew (Eds.), New 
York: Wiley, 1996, p. 132. 

18. E. E. Davis and M. F. Bowen, Sensory physiological basis for attraction in mosquitoes, J. Am. Mosq. 
Control Assoc., 10. 316, 1994. 

19. T. R. Burkot, Non random host selection by anopheline mosquitoes, Parasitol. Today, 4, 156, 1988. 

20. C. F. Curtis et al.. The relative efficacy of repellents against mosquito vectors of disease, Med. Vet. 
Entomol., 1, 109, 1987. 

21. R. De Jong and B. G. J. Knols, Selection of biting sites on man by two malaria mosquito species, 
Experientia, 51. 80. 1995. 

22. D. N. Diirrheim and J. M. Govere, Targeting vector behaviour and characteristics for effective malaria 
control in southern Africa, Appl. Env. Sci. Public Hlth., 1, 73, 2003. 

23. B. G. J. Knols, Odour-mediated host-seeking behaviour of the Afrotropical malaria vector An. 
gambiae Giles, Ph.D diss.. University of Wageningen, The Netherlands, 1996. 

24. A. J. Haddow, Observation on the biting habits of African mosquitoes in the genus Eretmapodites 
Theobold, Bull. Entomol. Res., 14, 761, 1956. 

25. M. T. Gillies. The role of carbon dioxide in host-finding of mosquitoes (Diptera: Culicidae): A review. 
Bull. Entomol. Res., 70, 525, 1980. 

26. B. G. J. Knols and R. De Jong. Limburger cheese as an attractant for the malaria mosquito An. gambiae 
s.s., Parasitol. Today, 12, 159, 1996. 

27. B. G. J. Knols, R. De Jong, and W. Takken, Differential attractiveness of isolated humans to 
mosquitoes in Tanzania, Trans. R. Soc. Trop. Med. Hyg., 89, 604, 1995. 

28. A. S. Self. M. H. M. Abdulcader, and M. M. Tun, Preferred biting sites of Culex pipiens fatigans on 
adult Burmese males. Bull. World Health Org., 40. 324-327, 1969. 

29. J. D. Charlwood, R. Paru, and H. Dagaro, Raised platforms reduce mosquito bites, Trans. R. Soc. Trop. 
Med. Hyg., 78, 141, 1984. 

30. J. S. Keystone, Of bites and body odour. Lancet, 347. 1423, 1996. 

31. L. E. O. Braack et al.. Biting pattern and host seeking behaviour of An. arabiensis (Diptera: Culicidae) 
in North-eastern South Africa, J. Med. Entomol, 31, 333, 1994. 

32. D. N. Durrheim and J. M. Govere. Malaria outbreak control in an African village by community 
application of “deet” mosquito repellent to ankles and feet, Med. Vet. Entomol., 16, 112, 2002. 

33. P. Granett, Studies of mosquito repellents I. Test procedure and method of evaluating test data, 
J. Econ. Entomol, 33, 563, 1940. 


© 2006 by Taylor & Francis Group, LLC 



Techniques for Evaluating Repellents 


159 


34. J. K. Trigg, Evaluation of a eucalyptus-based repellent against Anopheles spp. in Tanzania, J. Am. 
Mosq. Control Assoc., 12, 243, 1996. 

35. A. Bacot and G. Talbot, The comparative effectiveness of certain culicifuges under laboratory 
conditions. Parasitology, 11, 221, 1919. 

36. C. E. Schreck, N. Smith, and H. K. Gouck, Repellency of A,A-diethyl-m-toluamide (deet) and 
2-hydroxyethyl cyclohexanecarboxylate against the deer fly Cluysops flavidus Wiedemann. J. Med. 
Entomol., 13, 115. 1976. 

37. M. L. Fedder, On the method of laboratory testing of insect repellents for different insects, Med. 
Parazitol., 30, 730, 1961. 

38. A. A. Potapov, Olfactometer for repellent testing on fleas, Med. Parazitol., 1, 97, 1968. 

39. M. Bar-Zeev and S. Gothilf. Laboratory evaluation of tick repellents, J. Med. Entomol., 10, 71, 1973. 

40. C. E. Schreck, Techniques for the evaluation of insect repellents: A critical review, Annu. Rev. 
Entomol., 22, 101, 1977. 

41. H. H. Yap, K. Jahangir, and J. Zairi, Field efficacy of four insect repellent products against vector 
mosquitoes in a tropical environment, J. Am. Mosq. Control Assoc., 16, 241, 2000. 

42. C. F. Curtis and N. Hill, Comparison of methods of repelling mosquitoes, Entomol. Exp. Appl., 49, 
175, 1998. 

43. M. S. Fradin and J. F. Day, Comparative efficacy of insect repellents against mosquito bites, N. Engl. 
J. Med., 347. 13, 2002. 

44. B. G. J. Knols, R. De Jong, and W. Takken, Differential attractiveness of isolated humans to 
mosquitoes in Tanzania, Trans. R. Soc. Trop. Med. Hyg., 89, 604. 1995. 

45. R. I. Rose. Pesticides and Public Health: Integrated methods of mosquito management, Emerg. Infect. 
Dis., 7, 17, 2001. 

46. A. Seyoum et al., Traditional use of mosquito-repellent plants in western Kenya and their evaluation in 
semi-field experimental huts against Anopheles gambiae : Ethnobotanical studies and application by 
thermal expulsion and direct burning, Trans. R. Soc. Trop. Med. Hyg., 96, 225, 2002. 

47. J. M. Govere et al., Local plants as repellents against Anopheles arabiensis, in Mpumalanga Province, 
South Africa, Cent. Afr. J. Med., 46, 213, 2000. 


2006 by Taylor & Francis Group, LLC 



© 2006 by Taylor & Francis Group, LLC 



9 


Use of Olfactometers for Determining Attractants and 
Repellents 


Jerry F. Butler 


CONTENTS 

Introduction.161 

Review of Literature.162 

Olfactometer Test Systems.163 

Research Objectives.165 

Laboratory Olfactometer.165 

Summary of Olfactometer Development.167 

Summary of Olfactometer Market Sample Tests.168 

Market Sample Results.168 

Summary of Olfactometer Research.169 

References.191 


Introduction 

The development of attractants and repellents for future management of blood-feeding flies, mosquitoes, 
and other arthropods requires that behavioral regulating compounds (semiochemicals) first be identified. 
Preferably, these materials should be noninsecticidal, so that they may reduce selective resistance 
problems. When possible, semiochemical compounds should also be obtained from natural self- 
protective systems that are developed through plant and animal evolution, as these materials will be 
long term products of natural selection. These materials can be developed as attractants for baits and 
traps, as repellents for individual hosts, and as area treatments for exclusion of pests. They would most 
desirably be used to develop a push-pull system with attractant traps on the perimeter and repellents 
placed on or near the hosts to first capture and then exclude blood feeders. 

Identification procedures for repellent and attractant semiochemicals have historically been based on 
several approaches. “Trial and error” is the most common. In trial and error, test mixtures are placed into 
traps as baits or on skin to see if they will work. “Ask the insect to choose its preference” is a second 
approach. Here, semiochemicals within test systems where insects select or avoid points of treatment are 
exposed and the insects choose the semiochemical that they prefer. The third system is called “ask the 
insect to use its sensors to detect semiochemicals.” In this system, neural pathways to the insect’s brain 


161 


© 2006 by Taylor & Francis Group, LLC 

















162 


Insect Repellents: Principles , Methods, and Uses 


are probed to monitor detection. Neural activity reflects the insect both detecting and processing 
information and indicates the importance of the compound to the insect. However, monitoring neural 
activity requires skill in not only obtaining neural signals but also decoding their output to identify 
whether or not the signal has a positive or negative identity. Whichever system is used, it is imperative to 
make field evaluations that can determine the activity of semiochemicals under the proposed usage. 


Review of Literature 

Arthropods’ responses to semiochemical treatments can be identified based on physiological processes 
that are triggered by chemoreception and result in either attractancy or repellency behavior. 1-4 Chemical 
cues may be derived from food sources found in host plants or animals; larval habitats such as decaying 
organic material; oviposition sites in manure, soil, or water; and from the insects themselves. Selection 
by the arthropod requires that they detect key elements by olfactory senses, located on the antennae and 
maxillary palps, or visual senses. 5-8 

The complex behavioral sequence that results in host seeking and host-location by hematophagous 
arthropods involves an array of both chemical and physical cues. 9-16 These clues originate from the 
environment and are modified by visual cues, thermal effects, air movement, relative humidity, and 
chemical stimuli. 2,10 ' 14 ' 17-25 Host location at a distance is thought to be regulated by stimuli generated by 
the host, such as an increase in temperature (detected as infrared radiation), carbon dioxide (C02), 
and other host generated gasesr As the biting arthropod approaches the host, such stimuli orient 
the arthropod to the final landing site. 16 ' 27-31 Vision has also been found to be important in host 
location. 15 ' 21-23 ' 32 ^ 10 

Presumably, insects detect odors because odors cause changes in the electrical activity of primary 
olfactory receptor neurons contained within the antenna and maxillary palpal sensillae. Such nervous 
signals can be measured in the laboratory at a cellular level using probes placed at the neural receptors 
and the olfactory lobe of the insect’s brain. 7 ' 8 ' 41 

In blood-feeding insects, olfactory components in host finding are regulated in part by the plume of 
C02 that serves as a primary attractant along with other gases, including blood gases. 14 ’ 29 ’ 42 

Host location for parasitic species is complex. It requires that the parasite integrate all of the arthropod 
senses while it simultaneously gathers momentum during the process of host location. 26 The sequences 
of behaviors involved in host searching are susceptible to manipulation or interference by humans. 43 ' 44 
For example, host odor and the odors of excretory products have been found to be highly attractive to 
tsetse flies (Glossina spp.). 4 7 Takken 14 reviewed the active odors influencing host location by 

mosquitoes. Whether for mosquitoes or tsetse, identification of the components of such odors has 
led to the development of effective baits for sampling or controlling populations 48 The chemical, 
l-octen-3-ol, isolated from cattle (ox breath), has also been found to be a potent olfactory stimulant and 

r i • 49-51 

attractant tor tsetse and mosquitoes. 

Krijgman, 52 conducting experiments with the stable fly, Stomoxys calcitrans , reported orientation in a 
simple olfactometer to the odor of fresh blood. 53 Tests of various components and fractions of blood as 
attractants resulted in the discovery of an extremely volatile attractant constituent for Culex mosquitoes 
and Stomoxys. It is believed that this volatile fraction of blood diffuses through the skin of the host and is 
an important factor in attracting mosquitoes and biting flies to the host. 54 McKenzie 16 demonstrated that 
materials collected from human skin were attractive to host-seeking mosquitoes. 

Among the olfactory stimuli implicated in host location to date are carbon dioxide, lactic acid, acetone, 
butanone, octenol. phenolic components of urine, and oils on the skin. 14 ’ 16 ’ 28 ’ 29 ’ 55 ’ 56 Similar kinds of 
materials are probable cues utilized for finding oviposition sites by house flies, stable flies, and horn flies 
and include products of decaying plant and animal matter. 


© 2006 by Taylor & Francis Group, LLC 



Use of Olfactometers for Determining Attractants and Repellents 


163 


Olfactometer Test Systems 

Searching for semiochemicals that are detectable by blood-feeding flies and mosquitoes requires “asking 
the insect to choose its preference” to identify their detection of odors and components that can be 
isolated from the environment. This entails presenting choices to insects within test systems that isolate 
identified cues. 24,25 


Exhaust fan 


~¥~ 

I 



FIGURE 9.1 (a) Laboratory olfactometer top-view patent drawing, (b) Laboratory olfactometer side-view drawing 

(U.S. Patent 4,759,228). 


© 2006 by Taylor & Francis Group, LLC 

































































































164 


Insect Repellents: Principles, Methods, and Uses 



WmZZ2ZZZZ^Z2ZZ2ZZZm^m 



Ll 


111 


111 ill 


Light 

source 


Electric 
power for 
light source 





-4- 


►-f 


- 4 - 

- 4 - 


Air supply 
source 


-+- 


TlT TlTT 


Electic bias 
for pseudo host 


ITT 


Differential 

amplifier 


Program 
digital storage 


r 

Multichannel 

A-C Convertor 


r 

Printer 


r 

Computer 




(b) 


FIGURE 9.1 Continued. 


© 2006 by Taylor & Francis Group, LLC 


































































































































Use of Olfactometers for Determining Attractants and Repellents 


165 


Olfactometer test systems, effectively developed by Dethier, use the insect’s behavioral choices to 
place numbers on preference activities. 3 Optimum test systems present common sources of temperature, 
relative humidity, light, air quality, air flow, and when possible include the use of insect behavioral 
geotaxis to aid in triggering choices. The test cells are adjusted to allow free behavioral choices for the 
test species. Initial test chambers were observable, simple activity chambers with air flows, in which 
mosquitoes and other insects were allowed to move onto exposed, treated surfaces or host arms. 3 ' 24 ' 57-59 
A U.S. Department of Agriculture (USDA) olfactometer at the Mosquito and Vector Research 
Laboratory, Gainesville, Florida, U.S.A., based on the Dethier 3 model, has been in use as a screening 
device for many years. 56,60 

A two-choice “Y” tube olfactometer has been used to identify the pheromonal activity for horn and 
stable flies. It offers insects two choices. 61-63 This system has been further developed into our current test 
system with choices presented like wedges in a pie (hereafter: pie-type) for evaluating the response of 
arthropods to multiple odors or choices on treated artificial hosts. 24,25 


Research Objectives 

Our research objectives were to develop an arthropod olfactometer for rapid screening of semiochem- 
icals against a wide range of insects, mites, and ticks to evaluate the responses of arthopods to certain 
attractants, repellents, feeding stimuli, and oviposition stimuli. The system was adapted to measure 
arthropod preference by mapping arthropod movement in relation to a treated air stream, a treated 
surface, or a suitable food or oviposition source. The insect was then used to select potential compounds 
with strong semiochemical activity. 


Laboratory Olfactometer 

A multiport pie-type olfactometer was developed to electronically quantify insect feeding activity on ten 
compounds simultaneously for a set time period. 25,64-67 This olfactometer was developed to rapidly 
screen and compare large numbers of materials through a supporting grant from International Flavors and 
Fragrances, Inc. (IFF, Union Beach, New Jersey, U.S.A.). The multiport system was made possible by 
recent technological advancement in electronics and computers that are capable of both regulating test 
conditions and data-logging the results. 24,25,68 

The olfactometer integrated both computer and electronic detection to measure insect position and the 
act of feeding within a pie-type choice chamber. This design allowed the olfactometer to present testable 
materials and measure arthropod response to multiple chemical stimuli, with electronic monitoring of 
insect choice contact for eight hours or more. Comparisons were made by offering individual insects up 
to ten different choices presented simultaneously on ten artificial hosts or air streams (Figure 9.1a, b and 
Figure 9.2). 

The artificial hosts were made up of an agar base containing normal saline, feeding attractants, and 
cow blood covered by a silicone membrane. 16,25,69,70 This artificial host was attractive to blood-feeding 
insects, but they were unable to complete feeding due to the gel media. They continued to attempt 
feeding for more than eight hours. 16 The olfactometer created distinct and contiguous odor fields that 
could be easily entered, left, and reentered by the arthropods seeking a source. An electrical signal 
generated by the insect contacted with the gel medium was differentially amplified up to 10,000 times 
and data logged to a computer file for analysis. The feeding contact signal amplification with the 
differential amplifiers required the olfactometer to be housed in a temperature controlled, light-proof, 
Faraday-cage room (Lindgren Enclosures, Model No. 18-3/5-1, Glendale Heights, Illinois, U.S.A.) to 


© 2006 by Taylor & Francis Group, LLC 






166 


Insect Repellents: Principles, Methods, and Uses 




FIGURE 9.2 University of Florida multiport olfactometer in Faraday cage: paired T configuration (top), open 10-port 
configuration (bottom). 


© 2006 by Taylor & Francis Group, LLC 






Use of Olfactometers for Determining Attractants and Repellents 


167 


control extraneous electrical noise and test parameters. 16,25,66,69 This was required to maintain an 
acceptable signal to noise ratio (Figure 9.1a,b, and Figure 9.2). 

The air supply was augmented with moist C0 2 that was introduced into the air-stream by a computer- 
controlled valve at a rate of 250 mL/min (4 min on and 6 min off). This was used to expose treatments 
and mosquitoes to expected C0 2 near the artificial host, in case host C0 2 production affected the material 
being evaluated. The comparisons were made under identical light, temperature, air flow, and relative 
humidity, with exposure for up to 8 h so changes in activity and treatment could be monitored. 

The large data sets were manipulated with computer programs specifically written for processing these 
data. The activity was recorded as feeding contact, summed as bite-second totals on 10-min or 1-h 
intervals, and data-logged via computer. The test series were randomized and replicated over time for 
statistical analysis, with comparisons made to standard attractants and repellents within each trial. 
Optimum trials included an untreated standard and both attractants and repellents to statistically separate 
choices. The trials were conducted as replicated tests with analysis of variance utilized to evaluate the 
overall test significance. The significant tests (P<0.05) were then compared at a fixed time interval for 
differences within the trial. Data was summarized by time intervals of 4 or 8 h or as an 8-h time series of 
activity in a raster graph to show treatment formulation change over time. 

Standard statistical analysis was designed as randomized multi-choice tests, with ten randomly placed 
artificial hosts. Software designed to log data on contact seconds of feeding and biting (Medusa 2.1.2 
F&B designed by Nick Hostettler, Gainesville, Florida, U.S.A.) was used to consolidate and analyze the 
number of bite-seconds per sample over an 8-h period. Normalized data were compared to the standard 
using a one-tailed t test to determine whether there was any significant difference between samples. 
ANOVA was used to separate interaction and independent error term to avoid misrepresenting actual 
significance. Data were normalized by transformation using the square root of (n + 1). 

The trial consisted of 8-h replications as noted in Figure 9.3 through Figure 9.8. The replications 
were used in the final statistical analysis if they had no mechanical, electrical, or behavioral 
complications (e.g., amplifiers not communicating with the computer, sensors shorting, insects 
trapped behind sensors, no activity recorded for the standard untreated host during a replication). 


Summary of Olfactometer Development 

The laboratory olfactometer trials were conducted on a large number (greater than 3,999) of 
semiochemicals supplied by the granting agency, International Flavors and Fragrances, and other 
sources. The majority of tests were conducted on Aedes aegypti mosquitoes, house flies (Musca 
domestica), and horn flies (Haematobia irritans). The semiochemicals were principally selected extracts 
of plants and animals similar to those in Mookherjee et al. 24 Seventy-seven patents have been issued on 
attractants, repellents and test equipment as a result of this project. 

Table 9.1 lists the number of tests conducted. Table 9.2 the materials patented as attractants, and 
Table 9.3 the materials patented as repellents from the system through 1999. This list includes 28 
attractants and 89 repellents that were “new to science.” It should be noted that some of these 
compounds are dosage dependent. Deet was listed here as an attractant, although it is presented in the 
literature as a repellent. Deet is actually considered an inhibitor. It works by inhibiting the mosquito’s 
ability to sense L-lactic acid, effectively blocking antennal receptors. However, there are studies that 
suggest that at some levels, deet appears to act as an attractant instead of an inhibitor. 71 Acting as an 
activator, deet increased mosquito and house fly catch rates when added at low rates to attractant bait 
traps. At high rates (0.005 g/cm 2 ) for house flies and mosquitoes in the olfactometer, it acted as a 
repellent; at low rates (less than 0.0025 g/cm 2 ), it was often seen as an attractant. 72-74 Our data 
indicated that this and some other materials were variable depending on the concentration, reversing 
the insect biological response with either high or low concentrations. 


© 2006 by Taylor & Francis Group, LLC 




168 


Insect Repellents: Principles, Methods, and Uses 


2500 



FIGURE 9.3 Horn fly (Haematobia irritans ) feeding response to artificial host skin treatment of 0.005 g with market 
sample repellents and conditioners (5 rep, 8-h consolidated-feeding assay). 


Summary of Olfactometer Market Sample Tests 

The market sample repellents and skin conditioners were obtained from commercial sources. These 
market samples (Table 9.4) were evaluated in the multiport olfactometer to determine their comparative 
value in protecting the artificial host from arthropod bites. Applied directly to the silicone membrane of 
the artificial host or volatilized in an air stream directly above the membrane, these were evaluated by 
summing the insect feeding activity for either 4 or 8 h. Data were normalized using the square root of 
(n +1). Evaluations were also made as a time series to determine changes in activity over the test period. 
Formulations as tested are presented in Table 9.4. These were evaluated against horn flies (Haematobia 
irritans), stable flies (Stomoxys calcitrans), and the yellow fever mosquito (Aedes aegypti). Other 
mosquitoes were evaluated against herbal repellents compared to 6.5% deet products. 


Market Sample Results 

The mean comparisons for market sample tests are presented to give a general relationship to the activity 
when compared to the untreated hosts (code #3776 std host). The three most effective market sample 


© 2006 by Taylor & Francis Group, LLC 

















Use of Olfactometers for Determining Attractants and Repellents 


169 


1200 



I I I I I I I I I I 


o 

O 

0) 

a 

O) 

tt> 

C/3 O 

> 

§ 


Q 

CD 

0 

"O 

0 

(D 

(/ ) 

ZT 0 

O’ 

03 

0 


03 

3 

CD 

03 

V< 

0 

U) 

CD 

O 0 

C/3 

o 

0 

O’ 

0 

0 

d 

GO 

vO 

0 s 


ro 

O) 

xp 

0 s 

0 

■p 

O’ 

0 

8 ? 
o g 

O’ 

o' 

0 

O’ 

03 

O’ 

03_ 

O 

O’ 

03 

o 






o 

2 c£ 

c_ 

0' 




r 5 c n 

® ® 


FIGURE 9.4 Stable fly (Stomoxys calcitrans) feeding response to artificial host skin treatment of 0.005 g with market 
sample repellents and conditioners (6 rep, 8-h consolidated-feeding assay). 


repellents in these trials on horn flies were #3902 UltraThon™ (31.5% deet), #3903 Cutter® (21% deet), 
and #3907 Absorbine Jr.® (Figure 9.3). The four most effective repellents on stable flies were #3905 
Skin-So-Soft® Bath Oil, #3909 Green Ban® (10% citronella, 2% peppermint, and other plant extracts), 
#3902 Ultrathon, and #3907 Absorbine Jr. (Figure 9.4). The three most effective repellents on Aedes 
aegypti were #3902 Ultrathon, #3909 Green Ban, and #3905 Skin-So-Soft Bath Oil (Figure 9.5). 
Additional trials were conducted comparing the present market standard OFF! Skintastic™ 
Insect Repellent (6.65% deet) to the herbal repellents MosquitoSafe™ and TickSafe™ (30% geraniol). 
In all trials when geraniol based herbal repellents were tested with various arthropod species, geraniol 
based repellents had equal or significantly better repellent activity as skin treatments (Figure 9.6 through 
Figure 9.8). 


Summary of Olfactometer Research 

A laboratory olfactometer developed at the University of Florida is a system that electronically 
monitors treated artificial hosts or air-streams, with the choices arranged like wedges in a pie. It can 
be partitioned as a one, two, five or ten-choice system so comparative samples can be evaluated. 
The olfactometer can also be configured to evaluate dosage levels of the same compounds so that 
threshold activities can be evaluated for materials that change from attractants at low concentrations 


Ultrathon is a registered trademark of 3M Corp., Minneapolis, MN; Cutter is a registered trademark of United Industries 
Corp., St. Louis, MO; Absorbine Jr. is a registered trademark of W.F. Young, Inc., East Longmeadow, MA; Skin-So-Soft is a 
registered trademark of Avon Products, Inc., New York, NY; Green Ban is a registered trademark of Mulgum Hollow Farm, 
Brookvale, New South Wales, Australia; OFF! Skintastic is a registered trademark of S.C. Johnson and Son, Inc., Racine, WI. 


© 2006 by Taylor & Francis Group, LLC 




















170 


Insect Repellents: Principles, Methods, and Uses 


250 



o' 


FIGURE 9.5 Aedes aegypti air-skin comparisons as bite second counts on treated artificial host skin (S) or air (A) at 0.005 g 
(8-h exposure). 



FIGURE 9.6 The mosquito {Aedes aegypti ) mean 8-h exposure as bite second response to treated artificial host skin (S) at 
0.002 g or air (A) at 0.005 g; results as average sqrt(2) of bite sec/h. 


© 2006 by Taylor & Francis Group, LLC 







2006 by Taylor & Francis Group, LLC 













172 


Insect Repellents: Principles, Methods, and Uses 


TABLE 9.1 

Insects Studied and Number of Semiochemicals Evaluated in the Patented 
Laboratory Multiport Olfactometer 


Insect 

Number of 

Tests 

Number of 

Attractants 

Number of 
Repellents 

Mosquito 

5,626 

189 

688 

House Fly 

1,400 

173 

423 

Horn Fly 

568 

28 

89 

# Patents Issued 

77 

28 

89 

# Compounds 


32 

120 


Source'. From B. D. Mookherjee et al.. Bioactive Volatile Compounds from Plants; 203rd 
National Meeting of the American Chemical Society, R. Teransishi, R. G. Buttery, and 
H. Sugisawa (Eds.), ACS Symposium Series, Vol. 525, Washington, DC: American Chemical 
Society, 1993, p. 35: J. F. Butler and J. S. Okine, Nuisance Concerns in Animal Manure 
Management: Odors and Flies, Proceedings of a Workshop, H. H. Van Horn (Ed.), Vol. 117, 
Gainesville, FL: University of Florida and Georgia Agriculture Cooperative Extension 
Station, 1995. p. 1. 


to repellents at high concentrations. The components of repellent formulations were evaluated with 
the system to determine their compatibility and loss rates over eight-hours time series evaluations. 
The system can also be configured to evaluate insect response to light wave (color) and frequencies 
of light (Hz). Arthropod activity is detected by touch or feeding contact sensors using differential 
amplifiers that increase the signal by 10,000 times so that contact seconds of activity can be data 
logged to a computer for analysis. The assays of first to tenth bite time intervals can also be 
determined with selected computer programs. Measurements are obtained as feeding or contact 
response in time and summed as contact bite-seconds/hour. Replications are made with treatment 
positions randomized within each trial to eliminate position effect and determine interaction that 
may occur. Overall, treatment significance is evaluated with comparisons between choices for the 
sum of activity for an eight-hours exposure. Individual one-tailed t tests are used to determine 
significant differences between two choice treatments. 

The laboratory olfactometer has been used as a rapid screening system of semiochemicals against 
a wide range of arthropods to evaluate their potential activity as attractants, repellents, and feeding 
stimuli. “New to science” repellents and attractants were identified. The semiochemicals with 
attractant and repellent activities were identified as new to science with 77 patents issued to date 
covering a total of 139 compounds. Presented in this chapter are the new-to-science attractants (33) 
and repellents (87) that have been identified and patented at the University of Florida using this test 
system. The standard market sample repellents were compared demonstrating the effectiveness of 
high rate deet products compared to low rate deet standards. When herbal repellents (geraniol) were 
compared to the lower rate deet products, several formulations demonstrated repellent activity equal 
to or better than the deet products. Research is underway to evaluate factors of human attractant 
and repellent activities. The results have demonstrated individuals with repellent and attractant 
characteristics. 16 The laboratory olfactometer was adapted to measure fly, mosquito, cockroach, tick, 
and flea responses to treated hosts or air streams. The tested mosquitoes and fly species evaluated in 
the olfactometer include mosquitoes ( Culex spp.. Anopheles quadrimaculatus, Aedes aegypti), the 
horn fly ( Haematobia irritans), the house fly (Musca domestica), and the stable fly ( Stomoxys 
calcitrans). 


© 2006 by Taylor & Francis Group, LLC 



Use of Olfactometers for Determining Attractants and Repellents 


173 


TABLE 9.2 

Attractants Identified and Patented from the Olfaction Research Program at the University of Florida, 
Including 32 Attractant Compounds Cited in U.S. and Foreign Patents 


1 -(2-Butenoyl)-2,6,6-trimethyl-1-1 -3-cycohexadiene 


2,3-Dimethyl-3-hexanol 


2-Methyl-3-pentenoic acid 





3-Methyl-3-buten-1 -ol 



3 -Ethyl-3 -hexanol 



3 -Ethy l-2-methyl-3 -pentanol 



9-Decen-l-ol having the structure 


10-Undecen-l-ol having the structure 


Alpha-damascone 



(continued) 


2006 by Taylor & Francis Group, LLC 




174 


Insect Repellents: Principles, Methods, and Uses 


TABLE 9.2 (continued) 


Alpha-terpineol 



Aryl moiety compound (a) 



Aryl moiety compound (b) 



Benzyl formate 



Beta-damascone 



d-Carvone 



Dibutyl succinate 


O 



O 


(continued) 


© 2006 by Taylor & Francis Group, LLC 



















Use of Olfactometers for Determining Attractants and Repellents 


175 


TABLE 9.2 (continued) 


(/-Limonene structure 



Dimethyl disulfide 


CH 3 

S 

\ 


ch 3 


Dimethyl substituted oxymethyl cyclohexene (la) 



d-Pulegone 



Ethyl ester of 2-methyl-3-pentenoic acid 



Isobutyric acid 


O 



OH 


Jasmine absolute; racemic borneol 



OH 


(continued) 


2006 by Taylor & Francis Group, LLC 



176 


Insect Repellents: Principles, Methods, and Uses 


TABLE 9.2 (continued) 


Marigold absolute 



Methyl-isoeugenol 







n-Dodecanol 



A/^V-Diethyl-ra-toluamide 



rraw^rra/is-Delta-damascone 



© 2006 by Taylor & Francis Group, LLC 



Use of Olfactometers for Determining Attractants and Repellents 


177 


TABLE 9.3 

Repellents Identified and Patented from the Olfaction Research Program at the University of Florida, 
Including 89 Repellents Compounds Cited in U.S. and Foreign Patents 


l-Octen-4-ol 


2,4-Dimethyl-4-phenyl-1 - 
butanol 





Acyclic and carbocyclic 
ketones, alcohols, alde¬ 
hydes, nitriles and esters 
and uses thereof (1) 


Acyclic and carbocyclic 
ketones, alcohols, alde¬ 
hydes, nitriles and esters 
and uses thereof (2.1) 




Acyclic and carbocyclic 
ketones, alcohols, alde¬ 
hydes, nitriles and esters 
and uses thereof (2.2) 


(continued) 



2006 by Taylor & Francis Group, LLC 



178 


Insect Repellents: Principles, Methods, and Uses 


TABLE 9.3 (continued) 


Acyclic and carbocyclic 
ketones, alcohols, alde¬ 
hydes, nitriles and esters 
and uses thereof (2.3) 



Acyclic and carbocyclic 
ketones, alcohols, alde¬ 
hydes, nitriles and esters 
and uses thereof (2.4) 



Acyclic and carbocyclic 
ketones, alcohols, alde¬ 
hydes, nitriles and esters 
and uses thereof (2.5) 


Acyclic and carbocyclic 
ketones, alcohols, alde¬ 
hydes, nitriles and esters 
and uses thereof (2.6) 



Acyclic and carbocyclic 
ketones, alcohols, alde¬ 
hydes, nitriles and esters 
and uses thereof (3) 


Alkyl cyclopentanone and 
phenyl alkanol derivative- 
containing composition (a) 




Alkyl cyclopentanone and 



© 2006 by Taylor & Francis Group, LLC 




Use of Olfactometers for Determining Attractants and Repellents 


179 


TABLE 9.3 (continued) 


Alkyl cyclopentanone, 
cycloalkanal and phenyl 
alkanol derivative- 
containing (b) 



Alkyl cyclopentanone, 
cycloalkanal 

and phenyl alkanol deriva¬ 
tive-containing (c) 



Alkyl cyclopentanone, 
cycloalkanal and phenyl 
alkanol derivative- 
containing (d) 


Alkyl cyclopentanone, 
cycloalkanal and phenyl 
alkanol derivative- 
containing (e) 




Bisabolene isomer (a) 



Bisabolene isomer (b) 




2006 by Taylor & Francis Group, LLC 



180 


Insect Repellents: Principles, Methods, and Uses 


TABLE 9.3 (continued) 


Bisabolene isomer (d) 


Bisabolene isomer (e) 


Bisabolene isomer (f) 


Bisabolene isomer (g) 


C12 branched alcohol 


C12 unsaturated ketone 
mixture 


Carbocyclic compounds 







(continued) 


© 2006 by Taylor & Francis Group, LLC 



Use of Olfactometers for Determining Attractants and Repellents 


181 


TABLE 9.3 (continued) 


Camekol dh structure 


Carbonate esters (a) 


Carbonate esters (b) 



Carbonate esters (c) 



Citronellol 


Cyclemone® (a) 


Cyclemone® (b) 





(continued) 


© 2006 by Taylor & Francis Group, LLC 






















182 


Insect Repellents: Principles, Methods, and Uses 


TABLE 9.3 (continued) 


Cyclemone® (c) 




Cycloalkanol derivative- 
containing composition (a) 


Cycloalkanol derivative- 
containing composition (b) 


Cycloalkanol derivative- 
containing composition (c) 


Cycloalkanol derivative- 
containing composition (a) 


Cycloalkanol derivative- 
containing composition (b) 


Cycloalkanol derivative- 
containing composition (c) 



Cycloalkanol derivative- 
containing composition (b) 



© 2006 by Taylor & Francis Group, LLC 




Use of Olfactometers for Determining Attractants and Repellents 


183 


TABLE 9.3 (continued) 


Cycloalkanol derivative- 
containing composition (c) 



and 



Cycloalkanol derivative- 
containing composition (d) 



Cycloalkanol derivative- 
containing composition 
(e2) 





Cycloalkanol derivative- 
containing composition (b) 


Cycloalkanol derivative- 
containing composition (c) 


Cycloalkanol derivative- 
containing composition (d) 



(continued) 


2006 by Taylor & Francis Group, LLC 



184 


Insect Repellents: Principles, Methods, and Uses 


TABLE 9.3 (continued) 


Cycloalkanol derivative- 
containing composition (e) 


Dihydrofloralol 



Dihydrofloralol 



Dimethyl substituted 
oxymethyl cyclohexane 
derivative (la) 



Dimethyl substituted 
oxymethyl cyclohexane 
derivative 




© 2006 by Taylor & Francis Group, LLC 



Use of Olfactometers for Determining Attractants and Repellents 


185 


TABLE 9.3 (continued) 


Geraniol precursor 


Geraniol 


Geraniol Coeur: nerol 


Geraniol Coeur: citronellol 


Geraniol Coeur: geraniol 


Hedione 








(continued) 


2006 by Taylor & Francis Group, LLC 



186 


Insect Repellents: Principles, Methods, and Uses 


TABLE 9.3 (continued) 


Isocyclogeraniol (a) 


Isocyclogeraniol (b) 



Karismal 




O 



O 


Ketones 


Ketones, aldehydes, and 
esters (a) 


Ketones, aldehydes, and 
esters (b) 



(continued) 


© 2006 by Taylor & Francis Group, LLC 



Use of Olfactometers for Determining Attractants and Repellents 


187 


TABLE 9.3 (continued) 


Ketones, aldehydes, and 
esters (c) 


Ketones, aldehydes, and 
esters (d) 


Ketone and Schiff base- 
containing compositions 
(a) 



OH 







Ketone and Schiff base- 



Kovone 


Kovone 



(continued) 


2006 by Taylor & Francis Group, LLC 



188 


Insect Repellents: Principles , Methods, and Uses 


TABLE 9.3 (continued) 


Lavonax 



Lyral (1) 


Lyral (2) 



Lyrame 



OH 




(continued) 


© 2006 by Taylor & Francis Group, LLC 





Use of Olfactometers for Determining Attractants and Repellents 


189 


TABLE 9.3 (continued) 


Melozone (a) 



Melozone (b) 


Melozone (c) 



Melozone, wherein 60-40 
mole percent is the 
compounds 






Nerol 


(continued) 



2006 by Taylor & Francis Group, LLC 



190 


Insect Repellents: Principles, Methods, and Uses 


TABLE 9.3 (continued) 


Orange flower ether 



OCH 3 


Organo-boron derivative 



Schiff base of ethyl vanillin 
and methyl anthranilate 



Substituted silane, 
digeranyloxy-dimethyl- 
silane structure 



© 2006 by Taylor & Francis Group, LLC 



Use of Olfactometers for Determining Attractants and Repellents 


191 


TABLE 9.4 


Market Sample Repellents, Skin Conditioners, and Herbal Based Repellents used in Olfactometer Trials 


Sample 

Name 

Manufacturer 

3776 

Standard Artificial Host Attractant 

N/A 

3902 

Ultrathon™ 3M EPA 58007-1 31.5% Deet 

3M St. Paul, MN 55144-1000 

3903 

Cutter 21% Deet EPA 121-129 

Miles Inc, Consumer Household Products Div. 
7123 W. 65th St., Chicago, IL 60638 

3904 

Skin So Soft Mosquito Flea and Deer Tick 
Repellent EPA 65233.1.806 

Avon Products, Inc. New York, N.Y. 10019 

3905 

Skin-So-Soft Bath Oil Spray 

Avon Products, Inc. New York, N.Y. 10019 

3906 

Off! Skintastic™ Insect Repellent 

S.C. Johnson & Son, Inc Racine, WI 53403 

3907 

Absorbine Jr.® Antiseptic Liniment 

W.F. Young, Inc. Springfield, MA 01103 

3908 

Water Babies® SPF 30 

Schering Plough Healthcare Products, Inc 
Memphis, TN 38151 

3909 

Green Ban Citronella With Calendula and Cajuput 

Mulgum Hollow Farm P.O. Box 225, Brookvale 
NSW 2100 Australia 

3910 

Bull Frog® Sunblock SPF 36 

Chattem, Inc. Chattanooga, TN 37409 

3911 

Quantum Buzz Away 

Quantum, Inc. P.O. Box 2791 Eugene, OR 97402 

3912 

TickSafe™ 

Naturale LTD (now Fasst Products Inc., Rockville 
Center, NY 11570) 9 Park Place, Great Neck, 
NY 11021 

3913 

MosquitoSafe™ 

Naturale LTD (now Fasst Products Inc., Rockville 
Center, NY 11570) 9 Park Place, Great Neck, 
NY 11021 

3923 

Alsenite® 

Imperial Builders Apopka, FL 

3928 

Experimental Attractant 

N/A 

3962 

Experimental Repellent 

N/A 

3964 

Repello Base with Deet 

Naturale LTD (now Fasst Products Inc., Rockville 
Center, NY 11570) 9 Park Place, Great Neck, 
NY 11021 

3965 

Experimental Base 

N/A 

3966 

Experimental Repellent 

N/A 

3967 

Experimental Repellent 

N/A 

3969 

Geraniol Wristband 

Naturale LTD (now Fasst Products Inc., Rockville 
Center, NY 11570) 9 Park Place, Great Neck, 
NY 11021 

3970 

Experimental Deet Wristband 

Naturale LTD (now Fasst Products Inc., Rockville 
Center, NY 11570) 9 Park Place, Great Neck, 
NY 11021 


References 

1. V. G. Dethier, Chemoreception, in Insect Physiology, D. Kenneth Roeder (Ed.), New York: Wiley, 
1953, p. 544. 

2. V. G. Dethier, The sensory physiology of blood-sucking arthropods, Exp. Parasitol. , 6, 68. 1957. 

3. V. G. Dethier, The Physiology of Insect Senses, New York: Wiley, 1963, p. 266. 

4. V. G. Dethier, The Hungry Fly: A Physiological Study of the Behavior Associated with Feeding, 
Cambridge: Harvard University Press, 1976. 

5. E. Stadler. Contact chemoreseption, in Chemical Ecology of Insects, W. J. Bell and T. Carde (Eds.), 
Sunderland. MA: Sinauer Associates, 1984. 

6. H. Mustaparta, Olfaction, in Chemical Ecology of Insects, W. J. Bell and T. Carde (Eds.), Sunderland, 
MA: Sinauer Associates, 1984. pp. 35-70. 


2006 by Taylor & Francis Group, LLC 




192 


Insect Repellents: Principles, Methods, and Uses 


7. N. E. Kiwia, Electrophysiological brain activity recorded from the house fly, Musca domestica (L.) 
(Diptera: Muscidae) stimulated by attractant and repellent semiochemicals, MS thesis. University of 
Florida, 1989, p. 126. 

8. A. J. Grant et al., Electrophysiological responses of receptor neurons in mosquito maxillary palp 
sensilla to carbon, J. Comp. Physiol. A., 177, 389, 1995. 

9. R. F. Chapman, Some experiments to determine the methods used in host finding in tsetse flies, 
Glossina medicorum. Bull. Entomol. Res., 52, 83, 1961. 

10. B. Hocking, Blood sucking behavior of terrestrial arthropods, Ann. Rev. Entomol., 16, 1, 1971. 

11. W. G. Friend and J. J. B. Smith, Factors affecting feeding by blood-sucking insects, Ann. Rev. 
Entomol., 22, 309, 1977. 

12. D. E. Kline, J. R. Wood, and J. A. Cornell, Evaluation of l-octen-3-ol as an attractant for Coquillettidia 
perturbans, Mansonia spp. and Culex spp. Associated with phosphate mining operations, J. Am. Mosq. 
Control Assoc., 6, 605, 1990. 

13. M. Jacobson, Glossary of Plant Derived Insect Deterrents, Boca Raton, FL: CRC Press, 1990, 
pp. 7-8. 

14. W. Takken, The role of olfaction in host-seeking of mosquitoes, Insect Sci. Applic., 12, 287, 1991. 

15. D. A. Burkett, Light color attraction and dietary sugar composition for several mosquito (Diptera: 
Culicidae) species found in north central Florida, Ph.D. diss., University of Florida, 139, 1998. 

16. K. E. McKenzie, Determining factors of preferential host selection by Aedes aegypti (Diptera: 
Culicidae), Ph.D diss.. University of Florida, 148, 2003. 

17. J. J. Laarman, The host seeking behavior of the malaria mosquito. Anopheles maculipennis atroparvus, 
Acta Leinden., 25, 1, 1975. 

18. A. W. A. Brown, The attraction of mosquitoes to hosts, J. Am. Med. Assoc., 196, 249, 1966. 

19. M. T. Gillies and T. J. Wilkes, A comparison of the range of attraction of animal baits and of carbon 
dioxide for some West African mosquitoes, Bull. Entomol Res., 59, 441, 1969. 

20. G. A. Vale, The flight of tsetse flies (Diptera: Glossinidae) to and from a stationary ox, Bull. Entomol 
Res., 67, 297, 1977. 

21. L. W. Dalton et al., Host location in the horn fly: Roles of heat, CCD, water vapor, cow produced odors in 
attraction, Southwest Entomol, 3, 147, 1978. 

22. H. G. Kinzer, J. M. Reeves, and J. W. Atmar, Host location by the horn fly: Field evaluation of an 
artificial devise for measuring attraction to various stimuli. Environ. Entomol, 7, 375, 1978. 

23. J. R. Agee and R. S. Patterson, Spectral sensitivity of stable, face, and horn flies and behavioral responses 
of stable flies to visual traps (Diptera, Muscidae), Environ. Entomol, 12, 1823, 1983. 

24. B. D. Mookherjee et al., Semio activity of flavor and fragrance molecules on various insect species, in 
Bioactive Volatile Compounds from Plants; 203rd National Meeting of the American Chemical Society, 
R. Teransishi, R. G. Buttery, and H. Sugisawa (Eds.), ACS Symposium Series, Vol. 525, Washington, 
DC: American Chemical Society, 1993, p. 35. 

25. J. F. Butler and J. S. Okine, Development of attractants and repellents for future management of pest fly 
populations, in Nuisance Concerns in Animal Manure Management: Odors and Flies, Proceedings of a 
Workshop, H. H. Van Horn (Ed.), Vol. 117, Gainesville, FL: University of Florida and Georgia 
Agriculture Cooperative Extension Station, 1995, p. 1. 

26. M. J. Lehane, Biology of Blood-Sucking Insects, London: Harper Collins Academic, 1991. 

27. W. Takken and D. L. Kline, Carbon dioxide and l-octen-3-ol as mosquito attractants, J. Am. Mosq. 
Control Assoc., 5, 311, 1989. 

28. J. F. Sutcliffe, Distance orientation of biting flies to their hosts. Insect Sci. Appl., 8, 611, 1887. 

29. L. P. M. Willemse and W. Takken, Odor-induced location in Tsetse flies (Diptera: Glossinidae), J. Med. 
Entomol., 31(6), 775, 1994. 

30. R. Garcia, Carbon dioxide as an attractant for certain ticks (Acarina: Argasidae and Ixodidae), Ann. 
Entomol Soc. Am., 55, 605, 1962. 

31. R. Garcia, Collection of Dermacentor andersonii (Stiles) with carbon dioxide and its application in 
studies of Colorado tick fever virus, Ann. J. Trop. Med. Hyg., 14, 1090, 1965. 

32. H. C. M. Parr, Studies of Stomoxys calcitrans (L.) in Uganda, East Africa. 2. Notes on life history and 
behavior, Bull Entomol Res., 53, 437, 1962. 


© 2006 by Taylor & Francis Group, LLC 



Use of Olfactometers for Determining Attractants and Repellents 


193 


33. J. Brady, The visual responsiveness of the tsetse fly Glossina morsitans Westw, (Glossinidae) to moving 
objects: The effects of hunger, sex, host odor and stimulus characteristics. Bull. Entomol. Res., 62, 257, 
1972. 

34. G. A. Vale, New field methods for studying the response of tsetse flies (Diptera: Glossinidae) to hosts. 
Bull. Entomol. Res., 64, 199, 1974. 

35. L. T. Hargett, Visual and olfactory responses of the horn fly, Haematobia irritans, Ph.D dissertation 
abstr., Oregon State University, 1962. 

36. L. T. Hargett and R. L. Goulding, Rearing the horn fly, Haematobia irritans, J. Econ. Entomol., 55,565, 
1962. 

37. E. G. Milstrey, Variation in horn fly, Haematobia irritans (L.) hydrocarbons and the effect of alternating 
pheromone levels in the field on wild horn fly population, 289. MS thesis. University of Florida, 1983. 

38. W. G. Bruce, The history and biology of the horn fly, Haematobia irritans (Linnaeus), with comments 
on control, N.C. Exp. Sta. Tech. Bull, 157, 1964. 

39. W. L. Bidlingmayer and D. G. Hem, The range of visual attraction and the effect of competitive 
attractants on mosquito (Diptera; Culicidae) flight. Bull. Entomol. Res., 70, 321, 1980. 

40. W. C. Smith and J. F. Butler, Ultrastructure of the Tabanidae compound eye: Unusual features for 
Diptera, J. Insect Physiol., 37(4), 287, 1991. 

41. A. J. Grant and R. J. O'Connell, Neurophysiological and morphological investigations of pheromone- 
sensitive sensilla on the antenna of male Trichoplusia ni, J. Insect Physiol., 32(6), 503, 1986. 

42. R. Galun, Diversity of Phagostimulants used for recognition of blood meal by haematophagus 
arthropods, in Host-Regulated Development Mechanisms in Vector Arthropods, D. Borovsky and 
A. Spielman (Eds.), Gainesville, FL: IF AS, 1986, p. 1. 

43. R. F. Harwood and M. T. James, Entomology in Human and Animal Health, New York: Macmillan, 
1979, p. 80. 

44. J. Colvin andG. Gibson, Host seeking behavior and management of Tsetse, Annu. Rev. Entomol., 37,21, 
1992. 

45. J. W. Hargrove and G. A. Vale, The effect of host odor concentrations on catches of tsetse flies 
(Glossinidae) and other Diptera in the field, Bull. Entomol. Res., 68, 607, 1978. 

46. M. L. A. Owaga, Preliminary observations on the efficacy of olfactory attractants derives from wild 
hosts of tsetse, Insect Sc. Appl., 5, 87, 1984. 

47. M. L. A. Owaga, Observations on the efficacy buffalo urine as a potent olfactory attractant for Glossina 
paJlidipes Austen, Insect Sc. Appl., 6, 561, 1985. 

48. G. A. Vale, S. Flint, and D. R. Hall, The field responses of tsetse flies, Glossina spp. (Diptera: 
Glossinidae), to odors of host residues, Bull. Entomol. Res., 76, 685, 1986. 

49. D. R. Hall et al., I-Octen-3-ol: A potent olfactory stimulant and attractant for tsetse flies isolated from 
cow odor. Insect Sc. Appl., 5, 335, 1984. 

50. D. E. Kline, Olfactory attractants for mosquito surveillance and control: l-octen-3-ol, J. Am. Mosq. 
Control Assoc., 10, 280, 1994. 

51. D. E. Kline et al., Field studies on the potential of butanone, carbon dioxide, honey extract, l-octen-3-ol, 
L-lactic acid and phenols as attractants for mosquitoes, Med. Vet. Entomol., 4, 383, 1990. 

52. B. J. Krijgsman, Reizphysiologische Untersuchungen an blutsaugenden Arthropoden im Zusammen- 
hang mit ihrer Nahrungswahl. I. Teil. Stomoxys calcitrans, Z. vergl. Physiol, 11, 702, 1930. 

53. A. G. Gatehouse and C. T. Lewis, Host location behavior of Stomoxys calcitrans, Entomol Exp. Appl., 
16, 275, 1973. 

54. R. Galun, The role of host blood in feeding behavior of Ectoparasites, in Dynamic Aspects of Host- 
Parasite Relationships, A. Zuckerman (Ed.), Vol. 2, New York: Wiley, 1975, p. 132. 

55. G. Nicolas and D. Sulilans, Immediate and latent effects of carbon dioxide on insects, Ann. Rev. 
Entomol, 34, 97, 1989. 

56. U. R. Bernier et al., Chemical analysis of human skin emanations: Comparison of volatiles from 
humans that differ in attraction of Aedes aegypti (Diptera: Culicidae), J. Am. Mosq. Control Assoc., 
18, 186, 2002. 

57. F. M. Feinsod and A. Spielman, An olfactometer for measuring host-seeking behavior of female 
Aedes aegypti (Diptera: Culicidae), J. Med. Entomol., 15, 282, 1979. 


2006 by Taylor & Francis Group, LLC 



194 


Insect Repellents: Principles, Methods, and Uses 


58. T. G. Floore et al., Mosquito trapping studies to determine the efficacy of two models of the 
Flowtron mosquito luring device, J. Fla. Anti-Mosq. Assoc., 56, 13, 1985. 

59. A. E. Eiras and P. C. Jepson, Responses of female Aedes aegypti (Diptera: Culicidae) to host odors 
and convection currents using an olfactometer bioassay. Bull Ent. Res., 84, 207, 1994. 

60. D. A. Carlson and C. D. Grant, Attraction of female mosquitoes ( Aedes aegypti and Anopheles 
quadrimaculatus Say) to stored human emanations: Effect of CCD H 2 0 and temperature adjustments. 
Proceedings and Papers of the Forty-seventh Annual Conference of the California Mosquito and 
Vector Control Association, Inc., Burlingame, CA: California Mosquito and Vector Control 
Association, 1979. 

61. H. T. Bolton, J. F. Butler, and D. A. Carlson, A mating stimulant pheromone of the horn fly, 
Haematobia irritans (L.) demonstration of biological activity in separated cuticular components, 
J. Chem. Ecol., 6, 951, 1981. 

62. S. Muhammed, J. F. Butler, and D. A. Carlson, Stable fly sex attractant and mating pheromones 
found in female body hydrocarbons, J. Chem. Ecol., 1, 387, 1975. 

63. J. W. Mackley, D. A. Carlson, and J. F. Butler, Identification of the cuticular hydrocarbons of the 
horn fly and assay for attraction, J. Chem. Ecol., 7, 669, 1981. 

64. J. F. Butler and I. Katz, Process for determination of repellency and attractancy. U.S. Patent 
4,759,228, 1987. 

65. A. B. Marin, C. B. Warren, and J. F. Butler. Method for repelling Aedes aegypti using 3, 7 dimethyl- 
6- octenenitrile and/or 2(3,3-dimethyl-2-norbornylidene) ethanol-1. U.S. Patent 5,734,892, 1991. 

66. R. A. Wilson, B. D. Mookheijee, and J. F. Butler, Electronic insect repellency and attractancy tester, 
U.S. Patent 5,134,892, 1991. 

67. J. S. Okine, Aspects of oogenesis in the horn fly, Haematobia irritans (Linnaeus) (Diptera: 
Muscidae), Ph.D diss.. University of Florida, 30, 1994. 

68. R. J. Symonds and D. M. Unwin, The use of a microcomputer to collect activity data, Physiol. 
Entomol., 7, 91, 1982. 

69. J. F. Butler et al.. In vitro feeding of Ornithodoros ticks for rearing and assessment of disease 
transmission, in Acarology VI, D. A. Griffiths and C. E. Bowman (Eds.), Vol. 2, Ellis Horwood: 
Chichester, 1984, p. 1075. 

70. E. L. Davis et al.. Laboratory blood-feeding of Culicoides mississippiensis Hoffman through a 
reinforced silicone membrane, J. Med. Entomol., 20, 177, 1983. 

71. E. J. Hoffmann and J. R. Miller. Reduction of mosquito attacks on a human subject by combination 
of wind and vapor-phase DEET repellent, J. Med. Entomol., 39, 935, 2002. 

72. R. A. Wilson, et al.. Use of 7V,/V-diethyl-M-toluamide and/or 2-methyl-3-pentenoic acid as insect 
attractants. U.S. Patent 4,876,087, 1989. 

73. R. A. Wilson, et al., Use of 7V,/V-diethyl-M-toluamide and/or 2-methyl-3-pentenoic acid as insect 
attractants. U.S. Patent 4.880,625, 1989. 

74. R. A. Wilson, et al., Use of W/V-diethyl-M-toluamide and/or 2-methyl-3-pentenoic acid as insect 
attractants. U.S. Patent 4,959,209, 1990. 


© 2006 by Taylor & Francis Group, LLC 



10 _ 

Discovery and Design of New Arthropod/Insect 
Repellents by Computer-Aided Molecular Modeling 


Raj K. Gupta and Apurba K. Bhattacharjee 

CONTENTS 

Background.195 

Historical Development of Arthropod Repellents.197 

Chemical Functional Requirements for Arthropod Repellent Compounds.198 

Electronic and Stereoelectronic Considerations.199 

Molecular Mechanism of Arthropod Repellent Activity.202 

Computational Procedure.204 

Results and Discussion.204 

Conformational Analysis.204 

Molecular Similarity Analysis of JH-Mimic and Deet Compounds.205 

Correlation of Molecular Orbital Properties in JH-Mimic and in Deet and Its Analogs ....208 

Molecular Electronic Properties of JH.209 

Development of a New Model for Repellent Research.209 

Chemical-Feature Based Considerations.209 

Significance and Uniqueness of the Methodology.210 

Computational Methods and Materials.210 

Procedure for Development of the 3D-QSAR Pharmacophore Model.210 

Bioassay for Mosquito Repellency.211 

Results and Discussion.212 

Concluding Remarks and Future Perspectives.221 

Acknowledgments.225 

References.225 


Background 

The goal of this chapter will be to focus on new, next-generation computer techniques of molecular 
modeling to illustrate to researchers in the field of arthropod repellents how information on the three- 
dimensional structure of small molecules can facilitate the identification, design, and synthesis of 
repellents. The emphasis is primarily on understanding the quantitative structure activity relationships 

195 


© 2006 by Taylor & Francis Group, LLC 



























196 


Insect Repellents: Principles , Methods, and Uses 


(QSAR) and mechanisms of action, enabling early planning for structural design, synthesis, and 
further development. 

Computer-assisted molecular modeling (CAMM) has been used to make remarkable advances in 
mechanistic drug design and in the discovery of new potential bioactive chemical entities in recent 
years. 1-3 CAMM techniques can provide five major types of information that are crucial for mechanistic 
design of drugs and potent new chemical compounds. They are: 

• Three-dimensional structure of a molecule 

• Chemical and physical characteristics of a molecule 

• Comparison of the structure of one molecule with other molecules 

• Graphical visualization of complexes formed between the modeled compound and proteins or 
other molecules 

• Predictions about how related molecules match the modeled ones, along with an estimate 
of potency 

With the advent of modern computers and graphic techniques, computations and visualization of 
structures ranging from small to large biomolecules, such as proteins, can be accomplished with greater 
speed and precision. The graphic tools in modern computers have made it possible not only to visualize 
the three-dimensional structures of large protein molecules, but also to perform interactive, virtual 
docking experiments between potential drug molecules and the binding sites of proteins. 

Molecular modeling has now become an inseparable part of research activities that require an 
understanding of molecular bases of environmental, biochemical, and biological processes. Compu¬ 
tational methodologies are routinely being used to make decisions about chemical development and also 
to perform direct experimental investigations. The current advances in these methodologies allow direct 
applications ranging from accurate ab initio quantum chemical calculations of stereoelectronic proper¬ 
ties, generation of three-dimensional pharmacophores, and performance of database searches to identify 
potent bioactive agents. 

Discovery of new insect repellent active ingredients is a complex process with ever-changing new 
technologies. For example, it still takes about 10 years and, on average, approximately $30 million to 
bring a new insect repellent to market. Thus, historically, any technology that can improve the efficiency 
of the process is highly valuable to the commercial industry. In silico technologies are relatively new and 
have shown remarkable success in recent years, particularly in virtual screening of compound databases. 
These technologies are primarily driven by both cost- and time-effectiveness of new active ingredient 
discovery. Although no model is perfect, regardless of whatever it represents, the ability to virtually 
screen hundreds of compounds in a few hours and to construct simulations of three-dimensional protein 
structures in a computer has pushed these technologies to the cutting edge of discovery of new insect 
repellent active ingredients. 

The ability of a bioactive molecule to interact with the recognition sites in receptors results from a 
combination of steric and electronic properties. Therefore, the study of stereoelectronic properties of 
these molecules can provide valuable information, not only to better understand the mechanism of action, 
but also to develop a reliable pharmacophore to aid in the design of more efficient analogues. Quantum 
chemical computations in modern computers can provide accurate estimates of the stereoelectronic 
properties of molecules, and thus can be used to assess interaction of potential repellent active 
ingredients with the receptor. 

Developing in silico three-dimensional pharmacophore models and using them selectively as 
templates for three-dimensional multi-conformer database searches to identify new potent compounds 
are a few of the many other remarkable successes of computational methodologies in recent years. 4 
A three-dimensional pharmacophore may be perceived as a geometric distribution of chemical features, 
such as a hydrogen bond acceptor, hydrogen bond donor, aliphatic and aromatic hydrophobic moieties, 
ring aromatic hydrophobicity, etc., in the three-dimensional space that defines the specific biological 


© 2006 by Taylor & Francis Group, LLC 



Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


197 


activity of a molecule. Pharmacophores are generated by multiple conformations from a set of 
structurally diverse molecules. The generated pharmacophore enables rapid screening of virtual 
molecules/libraries to identify potent and non-potent bioactive agents. 


Historical Development of Arthropod Repellents 

The occupational hazards of the military with respect to exposure to arthropod-borne diseases is in some 
ways very representative of the worst hazards presented to the public in any region. Military personnel 
generally come from outside the region and, therefore, may not have any more immunity to local 
pathogens than a newborn baby. Because of their extensive exposure outdoors during all times of day and 
in all kinds of weather, military personnel tend to receive the maximum number of arthropod bites. 
Lessons learned from the military experience with arthropod-borne diseases are, therefore, widely 
applicable to the public in general. 

Arthropods continue to be important to military operations when they act as vectors of disease. 5 
Arthropods serve as vectors in a number of different ways, from simple mechanical transmission of 
pathogenic organisms on the arthropod body—for instance, when house flies carry dysentery bacilli from 
infected feces to food—to the more complicated process of biological transmission, where the pathogens 
must spend part of their life cycle in the body of the arthropod before humans can be infected. 6-10 
Regardless of the specifics of the association, a vector is an organism that transmits a pathogen to a 
susceptible host. 1112 Arthropod-borne diseases are extensive both in terms of variety and public health 
impact, but few effective economical vaccines are currently available. 13-18 

The increase in the U.S. military's operations will continue to expose its personnel to region-specific 
biting arthropods and the vector-borne diseases that they carry. The degree of exposure will largely 
depend on environmental factors and operational intensity. Success of high-intensity held operations in 
regions of significant arthropod infestations may be associated with, or even depend on, a safe and 
effective repellent and service members’ adherence to its proper application. Because concerns have 
been raised in recent years regarding the safety of (V,(V-diethyl-m-toluamide (deet), one of the most 
widely used and reliable insect repellents available, the search for an alternative form of deet is also an 
important research goal for the U.S. Army. 19 

A fundamental activity of military medical entomologists is to establish the role that certain arthropod 
species or populations play in the transmission of a particular infectious disease to service members. 20 
Primary vectors are those that are mainly responsible for transmitting a pathogen to humans or animals; 
secondary vectors are those that play a supplementary role in transmission but would be unable to 
maintain disease transmission in the absence of the primary vector. 6 Mosquitoes are arthropods of special 
significance. They cause more human suffering than any other organism, with over two million people 
dying of mosquito-borne diseases every year. Not only can mosquitoes carry diseases that afflict humans, 
but they also transmit several diseases and parasites to which dogs and horses are very susceptible. These 
include dog heartworm, west Nile virus (WNV), and eastern equine encephalitis virus (EEEV). In 
addition, mosquito bites can cause severe skin irritation through an allergic reaction to the mosquito’s 
saliva. Mosquito-vectored pathogens include the protozoa that cause diseases such as malaria, nematodes 
that cause filarial diseases such as dog heartworm, and viruses that cause diseases such as dengue, many 
encephalitides, and yellow fever. 

Deet has been regarded as the standard mosquito repellent for the past several decades. However, as a 
repellent for human use, deet is not equally effective against all insects and arthropod vectors of 
diseases. 21-23 Furthermore, in most formulations, it has a short duration of action (no more than several 
hours) and several disagreeable cosmetic effects, such as unpleasant odor. Of greater concern is that 
when it is used in higher concentrations, the deeper skin penetration can cause potential toxicity. In 
addition, deet is a plasticizer that reacts with certain plastics and synthetic rubber. 


2006 by Taylor & Francis Group, LLC 



198 


Insect Repellents: Principles, Methods, and Uses 


With increased international travel, illnesses caused by mosquito-borne pathogens, such as malaria, 
yellow fever, dengue fever, filariasis, and viral encephalitis, are flaring up all over the globe. 19 One 
mosquito species that easily adapts to urban conditions, Culex pipiens, caused the epidemic of west Nile 
viral encephalitis in New York City in 1999 that has since spread up and down the eastern seaboard, 4 as 
well as the rest of North America. Insect repellents that are completely safe and more effective than 
current products would be important additions to the armamentarium of tools available to prevent 
transmission of arthropod-borne pathogens. 

Deet (the structural chemical name was previously MlV-diethyl-m-toluamide, but now designated N,N- 
diethyl-3-methylbenzamide) remains the gold standard of currently available insect repellents. This 
substance was discovered and developed by scientists at the U.S. Department of Agriculture in 1946 
during a program to develop better repellents for the U.S. Army. It was subsequently registered in 1957 
for use by the general public. It is a broad-spectrum repellent that is effective against mosquitoes, biting 
flies, chiggers, fleas, ticks, and other many other biting organisms. Twenty years of empirical testing of 
more than 20,000 other compounds has not resulted in another marketed chemical product with the 
duration of protection and broad-spectrum effectiveness of deet, 20 though recently introduced active 
ingredients may equal or exceed the effectiveness of deet (see Chapters 18-21). The U.S. Environmental 
Protection Agency (EPA) estimates that more than 38% of the U.S. population uses a deet-based insect 
repellent every year and that worldwide use exceeds 200,000,000 people annually. However, because it 
does not protect against all arthropod-borne diseases, a rational search for an alternative effective broad- 
spectrum repellent is needed. 

Despite the obvious desirability of finding an effective oral, systemic mosquito repellent, no such agent 
has been identified. 20 23 Thus, the search for the perfect topical insect repellent continues. This ideal 
agent would repel multiple species of biting arthropods, remain effective for at least eight hours, cause no 
irritation to the skin or mucous membranes, cause no systemic toxicity, resist abrasion and ruboff, and 
integrate into a greaseless and odorless formulation. 

Efforts to find such a compound have been hampered by the numerous variables that affect the inherent 
repellency of any chemical. Repellents do not all share a single mode of action, and surprisingly little is 
known about how repellents act on their targets. 22 Moreover, different species of mosquitoes may react 
differently to the same repellent. To be effective, a repellent must show an optimal degree of volatility, 
making it possible for an effective repellent vapor concentration to be maintained at the skin surface 
without evaporating so quickly that it loses its effectiveness. Many factors play a role in how effective 
any repellent is, including the frequency and uniformity of application, the number and species of the 
organisms attempting to bite, the user’s inherent attractiveness to blood-sucking arthropods, and the 
overall activity level of the potential host. 23 Abrasion from clothing, evaporation and absorption from 
the skin surface, wash-off from sweat or rain, higher temperatures, or a windy environment all decrease 
repellent effectiveness. 23 Each 10°C increase in temperature can lead to as much as a 50% reduction in 
protection time. The repellents currently available must be applied to all exposed areas of skin; 
unprotected skin a few centimeters away from a treated area can be attacked by hungry mosquitoes. 


Chemical Functional Requirements for Arthropod Repellent Compounds 

A number of studies have shown that chemical compounds containing specific functional groups or 
features are more effective arthropod repellents as measured by duration of protection. 6 Recently, we 
have reported a study 27 of similarity analysis of stereoelctronic properties (steric and intrinsic electronic 
properties) between natural insect juvenile hormone (JH), a synthetic insect juvenile hormone mimic 
(JH-mimic, undecen-2-yl carbamate), and deet and its analogues. Structure-activity studies on juvenile 
hormones have resulted in the discovery of JH-like compounds that mimic the morphogenetic activity of 


© 2006 by Taylor & Francis Group, LLC 




Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


199 


JH with the aim of controlling insect populations. However, no attempt has thus far been made to design 
insect repellents rationalizing the pharmacophores obtained from these studies. 

Understanding the mechanism of arthropod repellent activity is a major goal of chemists for designing 
more effective repellents. Because the biochemical steps leading to the desired repellent effect, especially 
the interaction with the three-dimensional molecular structure of the receptor(s), are still unknown, 
various efforts are being made to develop a general structural framework with high probability for 
repellent activity to guide the synthesis work. 28 The ability of the insect repellents to interact with the 
recognition sites in receptors results from a combination of steric and electronic properties. Therefore, 
the study of stereoelectronic properties of insect repellents can provide valuable information, not only to 
better understand the mechanism of repellent action, but also to develop a reliable pharmacophore to aid 
in the design of more efficient analogues. In addition, a three-dimensional (3D) pharmacophore model 
would be useful to identify the structural requirements for repellent activity that, in turn, could be utilized 
for 3D database queries to search for proprietary and/or commercially available compounds. 

Strategies for reducing the abundance and longevity of arthropod vectors of pathogens have been two¬ 
pronged, centering around habitat control (through chemical, physical, engineering, and biological 
means) and the use of personal protection in the form of insect or arthropod repellents. This chapter also 
reviews the quantitative structure activity relationships from currently available scientific data on 
synthetic and plant-derived insect repellents, and how new and effective repellents can be developed 
using computational methodologies. 

Few attempts have previously been made to apply QS AR modeling to repellent activities. This deficiency 
may be primarily due to availability of only semi-quantitative data on most of the extensive testing that was 
carried out earlier. 29 One of the first quantitative attempts for measuring molecular properties such as 
lipophilicity, vapor pressure, and molecular chain lengths was by Suryanarayana et al. 30 Working with 31 
insect repellent compounds, these researchers proposed a QSAR relationship in the form of 

PT = a log P + b log Vp + c log ML (10.1) 

where PT is the protection time provided by repellent activity, P is lipophilicity, Vp is vapor pressure, ML is 
molecular length, and a, b, c, and d are constants. 

Taking into account the paucity of quantitative data on insect repellents and the objectives discussed 
above, repellent structural and electronic properties were initially investigated using quantum-chemical 
methods to determine any functional dependence with protection time as measured by Suryanarayana 
et al. 30 The goal of their study was to provide predictive discriminators of insect repellency and a better 
understanding of the structure and repellency properties of these compounds. Although the authors’ 
initial study specifically addresses repellent efficacy, the technique of linking specific molecular 
electronic properties to biological activity is generally applicable to both efficacy and toxicity studies. 
The authors’ developmental model for structure-activity relationships and generation of pharmacophores 
was based on the following two approaches: 

• Consideration of electronic and stereoelectronic chemical properties of the known arthropod 
repellents to identify three-dimensional molecular-interaction pharmacophores. 

• Consideration of pharmacophores or chemical features of known arthropod repellents to 
identify three-dimensional pharmacophores with potential repellent activity. 


Electronic and Stereoelectronic Considerations 

Because physical-chemical properties of repellents play a significant role in their effectiveness, the role 
of molecular electronic properties in relation to repellent protection time was also assessed, using a series 
of deet analogues. 30 Using quantum chemical methods, lowest energy conformations and 


2006 by Taylor & Francis Group, LLC 



200 


Insect Repellents: Principles, Methods, and Uses 


molecular electronic properties were calculated for 31 amides divided into five different types: (1) N,N- 
dimethylamide, (2) IV.iV-diethylamide, (3) !V,lV-diisopropylamide, (4) IV-ethyl amides, and (5) piperidi- 
neamides (Table 10.1). Biological testing of the compounds was performed as reported by 
Suryanarayana et al. 0 Briefly, a dose of 1 mg/cnr was applied onto the external surface of a human 
fist, followed by exposure for 5 min to 200 female Aedes aegypti (aged 5-7 days). Exposure was repeated 
every 30 min until two consecutive bites were observed, defining the protection time as the time up to the 
period before the bites. 31 


TABLE 10.1 

Structure and Mosquito Repellent Protection Time of Deet and Its Analogs Organized According to Their 
Amide Substituents 


O O 




a = aromatic ring 


s 

= saturated ring 


Compound 

Structure 

Protection 

Time (h) 

Ring 

R 

Ri =Ri 

la 

oChlorobenzamide 

5 

a 

2-C1 

ch 3 

lb 

Cyclohexamide 

3 

s 

H 

ch 3 

lc 

m-Toluamide 

3 

a 

3-CH 3 

ch 3 

Id 

oEthoxylbenzamide 

2.83 

a 

2-OC 2 H 5 

ch 3 

le 

Benzamide 

1.67 

a 

H 

ch 3 

if 

p-Anisamide 

1 

a 

4 -OCH 3 

ch 3 

2a 

m-Toluamide 

5 

a 

3 -CH 3 

c 2 h 5 

2b 

Benzamide 

4 

a 

H 

c 2 h 5 

2c 

Cyclohexamide 

4 

s 

H 

c 2 h 5 

2d 

o-Ethoxylbenzamide 

3.5 

a 

2-OC 2 H 5 

c 2 h 5 

2e 

;?-Toluamide 

2.8 

a 

4 -CH 3 

c 2 h 5 

2f 

p-Anisamide 

1 

a 

4 -OCH 3 

c 2 h 5 

3a 

Benzamide 

3 

a 

H 

(C 3 h 7 

3b 

m-Toluamide 

2.67 

a 

3 -CH 3 

iC 3 H 7 

3c 

Cyclohexamide 

2 

s 

H 

i'C 3 H 7 

3d 

p-Anisamide 

1.17 

a 

4 -OCH 3 

iC 3 H 7 

3e 

oEthoxylbenzamide 

1.08 

a 

2-OC 2 H 5 

/c,h 7 

3f 

o-Chlorobenzamide 

1 

a 

2-C1 

iC 3 H 7 

3g 

p-Toluamide 

0.5 

a 

4 -CH 3 

i'C 3 H 7 

4a 

m-Toluamide 

0.67 

a 

3 -CH 3 

/?i r 2 

H C 2 H 5 

4b 

Benzamide 

0.58 

a 

H 

H C 2 H 5 

4c 

Cyclohexamide 

0.5 

s 

H 

H C 2 H 5 

4d 

/7-Toluamide 

0.08 

a 

4 -CH 3 

H C 2 H 5 

4e 

p-Anisamide 

0.08 

a 

4 -OCH 3 

H C 2 H 5 

4f 

oEthoxylbenzamide 

0.08 

a 

2-OC 2 H 5 

H C 2 H 5 

5a 

Benzamide 

3 

a 

H 

N, R„ R 2 
Piperidine 

5b 

Cyclohexamide 

2 

s 

H 

Piperidine 

5c 

m-Toluamide 

1.42 

a 

3 -CH 3 

Piperidine 

5d 

o-Chlorobenzamide 

1 

a 

2-C1 

Piperidine 

5e 

p-Toluamide 

1 

a 

4 -CH 3 

Piperidine 

5f 

p-Anisamide 

0.75 

a 

4 -OCH 3 

Piperidine 


© 2006 by Taylor & Francis Group, LLC 









Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


201 



FIGURE 10.1 (See color insert following page 204.) Optimized geometry and electrostatic potential profiles of three 
repellents that have good (top row, PT = 5h), moderate (middle row. PT = 1.4h) and poor (bottom row, PT = 0.08h) 
protection times. First column: optimized geometry; Second column: electrostatic potential onto surface of constant electron 
density (0.002 e/au 3 ); Third column: isoelectrostatic potential surface at — 10 kcal/mol. Atoms are colored black for carbon, 
red for oxygen, blue for nitrogen, and gray for hydrogen. The deepest blue surface in the second column is the most positive, 
and the deepest red surface is the most negative. (From D. Ma, K. Bhattacharjee, R. K. Gupta, and J. M. Karle, American 
Journal of Tropical Medicine and Hygine, 60, 1, 1999.) 

An examination of the electrostatic potential maps of the repellents at —10 kcal/mol (Figure 10.1), 
which roughly correspond to the electronic features beyond the van der Waals surface of the molecules, 
indicated that all repellents have a large extended negative potential region extending out from the 
carbonyl group. The electrostatic potential profiles of molecules are considered to be key features 
through which a molecule fits into a receptor at longer distances, and accordingly, promotes interaction 
between complementary sites with the receptor. 12 Although this potential characterizes the primary level 
of interaction with the receptor, there is no apparent relationship with the size or shape of these surfaces 
to protection time. Regions of positive potentials, the blue-colored regions in Figure 10.1, at the van der 
Waals surface indicate the electrophilic or acidic sites. Although the location of the most positive 
potential (deepest blue color) in the repellent molecules is found to be located adjacent to different 
hydrogen atoms on different molecules, the magnitude of the most positive potentials appears to be 
related to protection time. All compounds that provided protection for at least 2.8 h have a maximum 
positive potential in the range of 16.2-21.1 kcal/mol, whereas all compounds with a positive potential 
higher than 21.1 kcal/mol provided protection for no more than one hour. 33 Thus, the intrinsic 
electrophilicity of the repellent amides appears to play a role in the repellency of a compound. 

The dipole moment is another interesting electronic property that seems to have a role in repellent 
activity. This property is the intrinsic polarity of a molecule. Its magnitude is a good indicator of intrinsic 
lipophilicity or hydrophobicity. In general, the larger the magnitude, the more likely the compound is 
hydrophilic. In study conducted by the authors with 31 repellents, 33 the magnitude of the dipole moment 
for the most active repellents (PT>3.5 h) was found to be ranging between 3.25 and 3.82 Debye, an 


© 2006 by Taylor & Francis Group, LLC 


202 


Insect Repellents: Principles, Methods, and Uses 


indication that an optimal lipophilicity or hydrophobicity for this class of compounds is necessary for a 
molecule to be an active repellent. Although the orientation of the dipole moment of the repellents does 
not seem to have any link to protection time, the negative end of the dipole in these compounds was 
always observed to be pointing toward the oxygen atom of the carbonyl functional group. 

Atomic charges of the compounds seem to have a significant role in repellency. These charges indicate 
the intrinsic reactive character of the individual atoms constituting the molecules. The magnitude of 
negative charge on an atom characterizes the nucleophilic nature of the atom, whereas the magnitude of 
the positive charge correspondingly characterizes the electrophilic nature of the atom. In the data set 
from the above study of repellents, 33 a low atomic charge on the amide nitrogen atom in compounds 
having low PT values was observed. In general, it was observed that the more negative the charge on the 
amide nitrogen atom, the less protection time provided by the compound containing the atom. 

Molecular Mechanism of Arthropod Repellent Activity 

In the authors’ next study, 27 the stereoelectronic features of 15 of the 31 arthropod repellents (Table 10.2) 
reported by earlier workers were assessed, 30 by identifying both electronic and steric requirements for 
repellent activity. In addition, these profiles were compared with JH, not only to identify the 
stereoelectronic requirement, but also the probable mechanism of repellent action of the compounds. 


TABLE 10.2 


Structure and Mosquito Repellent Protection Time of Deet and Its Analogs Organized According to Their 
Amide Substituents 

O 



'R, 



,/Rt 


Compound 

Structure 

Protection Time 

(h) 

Ring 

R 

Rj —R 2 

la 

m-Toluamide 

(deet) 

5 

a 

3-CH 3 

c 2 h 5 

lb 

Cyclohexamide 

4 

s 

H 

C 2 H 5 

lc 

p-Anisamide 

1 

a 

4-OCH 3 

c 2 h 5 

2a 

o-Chlorobenza- 

mide 

5 

a 

2-C1 

ch 3 

2b 

m-Toluainide 

3 

a 

3-CHj 

ch 3 

2c 

p-Anisamide 

1 

a 

4 -OCH 3 

ch 3 

3a 

Benzamide 

3 

a 

H 

iC 3 H 7 

3b 

/7-Anisamide 

1.17 

a 

4 -OCH 3 

(C 3 H 7 

3c 

/7-Toluamide 

0.5 

a 

4 -CH 3 

*c 3 h 7 

R\ R 2 

4a 

m-Toluamide 

0.67 

a 

3 -CH 3 

H C 2 H 5 

4b 

Cyclohexamide 

0.5 

s 

H 

H C 2 H 5 

4c 

oEthoxylbenza- 

mide 

0.08 

a 

2 -OC 2 H 5 

H C 2 H 5 

N, R u R 2 

5a 

Benzamide 

3 

a 

H 

Piperidine 

5b 

m-Toluamide 

1.42 

a 

3 -CH 3 

Piperidine 

5c 

/7-Anisamide 

0.75 

a 

4 -OCH 3 

Piperidine 


© 2006 by Taylor & Francis Group, LLC 









Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


203 



(2) R = Et,R' = Me 

(3) R = R' = Me 



JH - mimic 

FIGURE 10.2 Structures of JH and JH-mimic. (From A. K. Bhattacharjee, R. K. Gupta, D. Ma, and J. M. Karle, Journal of 
Molecular Recognition, 13, 213, 2000.) 


Juvenile hormones (Figure 10.2) are ubiquitous growth regulators among insects and serve as a rational 
source for the design of synthetic insect growth regulators. 24 ’ 34 ’ 35 Structure-activity studies on JH have 
resulted in the discovery of JH-like compounds 25 ' 26 that mimic the morphogenic activity of the natural 
compound and are used commercially in insect control. However, to the knowledge of the authors, no 
attempt has been made thus far to design insect repellents based on JH chemical structure and activity. 

Although considerable research efforts have focused on why humans are attractive to insects, 36 
especially mosquitoes, and many chemicals have been discovered to have repellent activity, the mode of 
action of repellents remains poorly understood. In recent years, a tentative model of physical properties 
required for potent repellency of the two well-known insect repellents, 20 deet and A,)V-diethylpheny- 
lacetamide (DEPA), against Aedes aegypti mosquitoes has been proposed on the basis of their 
lipophilicity, vapor pressure, and molecular length. However, it is now widely believed 37-39 that a 
repellent must impact insects’ olfactory sense and that the olfactory sensation is primarily controlled by 
JH responses or activity. Therefore, an ideal repellent must be volatile, must come in contact with the 
mosquito’s olfactory organ, and have some degree of lipid solubility to trigger the olfactory sensation. 
The gas-phase molecular properties of the repellents thus should be an important aspect of studies on the 
mechanism of this interaction process with the olfactory organ. Quantum-chemically calculated 
stereoelectronic properties can provide an accurate estimate of gas-phase properties of molecules. 
Calculating and assessing these properties should be important objectives of chemists before the design 
and synthesis of new repellents. 

The mechanism of olfactory sensation may be viewed as an interaction of fundamental molecular 
forces between the repellents and the JH receptor of the insects, from the point of view of the century-old 
lock-and-key hypothesis of Emil Fischer. 40 According to the lock-and-key hypothesis, the biological 
activity of a compound may be accounted for through a molecular recognition mechanism between the 
biomolecule (lock) and the active compound (key). Because the JH receptor recognizes the stereo- 
electronic features of the active compound (repellent) that resemble those of JH itself, and not of atoms 
per se, 41-43 a comparative analysis of these features should provide a wealth of molecular level 
information that would not only aid in the design of new repellents, but also illuminate more completely 
the fundamental forces that affect the function and utility of the compounds. 

In recent years, several other studies based on quantitative derivations 44-46 have shown that this 
recognition process can be analyzed from three types of three-dimensional molecular similarity studies: 
(1) steric, (2) electrostatic, and (3) hydrophobic. It is well documented that bioactive compounds 
(ligands) will bind to a receptor in a similar manner by aligning their common molecular field or property 
characteristics to the receptor. 44 This concept is known as bioisosterism, wherein atoms or functional 
groups with similar properties are used for ligand design. 47 The study of bioisosterism has been one of the 


2006 by Taylor & Francis Group, LLC 



204 


Insect Repellents: Principles, Methods, and Uses 


most common means of discovery of new leads in pharmaceutical research. This method of selection of 
pharmacophores is mainly based on simple superposition principles using the analogy of complemen¬ 
tarity. The degree of complementarity between the molecular fields of the bioactive agent and its receptor 
should be directly related to the binding strength and relative activity of the agent. Similarity may also be 
determined by comparing the molecular graphs. 47 The word “similarity” in the present study means that 
two molecules have a common bioisosteric group. 

Accordingly, the authors’ study 27 assessed the similarity of the stereoelectronic properties of deet and 
its analogs to natural JH (Table 10.2, Figure 10.2 where R = methyl), and to a synthetic JH-mimic 
terpenoid. This study was an attempt to gain a better understanding of the mechanism of action of 
the deet-type insect repellents and to aid in the design and synthesis of more efficacious repellents. 
Undecen-2-yl carbamate, the JH-mimic, is a potent inhibitor of metamorphosis of the common mosquito, 
Culex pipens. 25 Structurally, deet, its analogs, and the JH-mimic all have an >N-C=0 fragment 
(Table 10.2, Figure 10.2), making it likely that a similar recognition interaction with the receptor will 
take place. Juvenile hormone is structurally different, containing an -0-C=0 fragment. However, 
similarity in molecular electronic shape, not solely the similarity in chemical structure, has long been 
recognized as the dominant factor for olfactory sensations. 48 Thus, using data from the earlier study 33 on 
predicting mosquito repellent activity from calculated stereoelectronic properties on 31 deet analogs, the 
authors carried out a computational study based mainly on the similarity analysis of the stereoelectronic 
properties of deet and 14 of its analogs with JH and JH-mimic using the semi-empirical AMI quantum 
chemical method. 

Computational Procedure 

Computational calculations were performed using SPARTAN version 5.0 49 running on a Silicon 
Graphics Indigo Extreme R4000 workstation. A detailed conformational search of JH and JH-mimic 
was performed by multiple rotation of single bonds in the compounds, thereby generating several 
low-energy conformers with varying population densities. The most abundant and the lowest energy 
conformers were identified. The geometry of these conformers was optimized, and the electronic 
properties were calculated using the optimized geometry. Geometry optimization and energy 
calculations were performed on the compounds in the gaseous phase at the semi-empirical AMI 
quantum chemical level using the method as implemented in SPARTAN. Three-dimensional 
molecular electrostatic potential (MEP) maps for all compounds were calculated using the 
SPARTAN calculations of the AMI-optimized geometry of the molecules. The MEPs were 
sampled over the entire accessible surface of a molecule (corresponding roughly to the van der 
Waals contact surface) and into space extending beyond the molecular surface, providing a measure 
of charge distribution from the point of view of an approaching reagent. The regions of negative 
potential indicated areas of excess negative charges and, therefore, suitable attraction sites in the 
molecule for the positively charged test probe. 

Results and Discussion 
Conformational Analysis 

The lowest energy conformers of JH-mimic were identified by systematic rotation of the single bonds. 
This procedure generated 256 conformers of JH-mimic, identifying the low energy conformers along 
with their corresponding Boltzmann population densities. The lowest energy conformer of JH-mimic has 
a 75.6% population density, whereas the other conformers were present in varying population densities 
ranging from 6 to 0.05%, with energies more than 5.0 kcal/mol greater than the lowest energy conformer. 
In the lowest energy conformers, the amide moiety in both JH-mimic and the deet compounds 33 was 
planar and superimposable on each other. The nonbonded distances, N-O and C=0-C R] , were within 
0.1 A of each other (Table 10.3). 


© 2006 by Taylor & Francis Group, LLC 


Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


205 


TABLE 10.3 


Selected Nonbonded Distances and Total Surface Areas Containing the C 7 , O, N, and Cri 
A toms 


Compound 

N-0=C, A 

C=0-C R i, A 

Surface Area, 

A 2 C 7 , O, N, Cri Atoms 

JH-mimic 

2.345 

2.871 

11.5 

la 

2.296 

2.751 

13.6 

lb 

2.282 

2.707 

13.6 

lc 

2.296 

2.746 

13.6 

2a 

2.293 

2.761 

13.6 

2b 

2.286 

2.749 

13.6 

2c 

2.285 

2.744 

13.6 

3a 

2.297 

2.810 

11.7 

3b 

2.296 

2.806 

11.7 

3c 

2.296 

2.809 

11.7 

4a 

2.296 

2.811 

11.7 

4b 

2.292 

2.815 

11.7 

4c 

2.298 

2.803 

11.7 

5a 

2.293 

2.761 

13.6 

5b 

2.293 

2.761 

13.6 

5c 

2.289 

2.757 

13.6 


Thus, although JH, JH-mimic, and the deet molecules have many degrees of conformational freedom 
in their structure, the main bioactive pharmacophore (the ester, carbamate, or the amide group) was 
found to be superimposable. This ensures the steric similarity of the pharmacophore . 27 

Conformational search calculations on the structure of the natural JH molecule where R = methyl 
identified three conformers of significant abundance, with a relatively small energy difference of 
3.9kcal/mol between the maximum and the minimum energy conformer. An energy barrier 
of 3.9 kcal/mol can easily be surmounted in biological systems, and, statistically, the distribution of 
all three conformers cannot be ruled out. The ester moiety is flat in all three conformers, and the 
nonbonded distances 0=0 and C=0-C R | are equal to 2.23 and 2.58 A, respectively. The three 
conformers differ in the conformation of the alkyl chain, due to rotations about the single bonds . 27 
Although JH has an epoxide moiety at one end of the molecule, this functionality does not appear to be 
important for growth regulator activity, because mimics lacking this functionality are potent growth 
inhibitors . 26 

Molecular Similarity Analysis of JH-Mimic and Deet Compounds 

The analysis of molecular recognition process in this investigation was based on the strategy of 
superimposition of the amide fragments and analysis of steric, electrostatic, and hydrophobic properties. 
JH-mimic is more structurally similar to the deet compounds than to the natural JH, as JH has an ester 
rather than an amide moiety. Therefore, this section concentrates on the comparison of the deet 
compounds to JH-mimic. 

The surface area and volume of the amide-containing portion of JH-mimic and the deet compounds 
show considerable similarity. The surface area containing the amide C 7 , O, N, and Cri atoms of 
JH-mimic and the deet compounds ranges from 11.5 to 13.6 A 2 (Table 10.3), whereas the calculated 
steric bulk of this amide portion in JH-mimic is 8.7 A 3 and the deet 0.2 A 3 compounds is 8.3 A 3 , 
respectively. Because steric complementarity is a prerequisite for ligand-receptor recognition, active 
biological agents with a common receptor binding site should possess sterically similar binding surfaces. 
This contribution reflects molecular size and overall shape, features important to the steric 


2006 by Taylor & Francis Group, LLC 



206 


Insect Repellents: Principles, Methods, and Uses 


complementarity of the ligand at the binding site. 50 Bond distances and angles of the amide moiety in 
deet and its analogs are also similar to those in the JH-mimic (Table 10.4). The bond distances of 
JH-mimic, deet, and deet’s most repellent analogs are within 0.007 A of each other. The bond angles and 
the dihedral angles differ from each other by up to 8°, and 7° to about 19°, respectively, a reasonable 
variation keeping in mind the large intrinsic differences in the geometry of the molecules. 

The similarity of electrostatic characteristics of the deet compounds with JH-mimic is likely to result 
in a similar recognition interaction with the JH receptor to promote binding interactions. The electrostatic 
characteristics include Mulliken charges, electrostatic potentials at essentially the van der Waals surface, 
dipole moment, and the profiles of electrostatic potential beyond the van der Waals surface. Electrostatic 
complementary interactions are believed to be long-range interactions between a ligand and its binding 
site, and are considered to be a very important contributing factor for a ligand/protein binding 
mechanisms. 5 '~ 54 This complementarity essentially means that the charge distribution of a substrate 
has to find its counterpart at the binding sites to allow maximum interaction with the receptor. 55 It works 
like a magnet between them and, thereby, contributes to the binding affinity. The calculated Mulliken 
charges and the electrostatic potential at the amide atoms in JH-mimic and deet and its analogs are 
presented in Table 10.5. 

The charge of the carbonyl oxygen atom of JH-mimic is — 0.02 electrons more negative than for deet 
and its analogs, while the negative potential by the carbonyl oxygen atom, —73.2 kcal/mol, falls in the 
same range as calculated for deet and its analogs, —73.1 to —77.3 kcal/mol. The carbonyl oxygen atom 
is also the site for the most negative potential 27 (see Figure 10.3) in deet, its analogs, and JH-mimic, 
making the carbonyl oxygen atom the most nucleophilic site in all the molecules, as this site has the 
maximum localized electron density. Although the calculated dipole moment is somewhat lower in 
JH-mimic than in the deet compounds, the dipole moment is pointing toward the carbonyl oxygen atom 
in all the compounds. Thus, the carbonyl oxygen atom seems to be the most reactive site in both 
JH-mimic and the deet compounds. 

Conversely, the site for the most positive potential is considered to be most electrophilic or acidic 
because of minimum electron density. In JH-mimic, the most positive potential is located by the amide 
hydrogen atom with a value of 37.3 kcal/mol, whereas in deet and its analogs, it is scattered around different 


TABLE 10.4 


Selected Structural Parameters of JH-Mimic and Deet Compounds 


Compound 

Bond Distance, A 

Bond Angle, 0 

Dihedral 
Angle, ° 
C r1 -N-C 7 = 0 

C R1 -N-C 7 -C 1 

or 

c r1 -n-c 7 -o 

c 7 =o 

N-C 7 

n-c r1 

n-c 7 =o 

Cr,-N-C 7 

JH-mimic 

1.241 

1.388 

1.442 

127.8 

119.1 

23.0 

-161.1 

la 

1.247 

1.392 

1.447 

120.8 

119.5 

2.6 

179.9 

lb 

1.248 

1.393 

1.448 

119.4 

118.5 

4.6 

-178.6 

lc 

1.247 

1.393 

1.447 

120.7 

120.7 

2.2 

179.5 

2a 

1.246 

1.387 

1.437 

120.9 

120.2 

7.8 

-174.9 

2b 

1.248 

1.387 

1.436 

120.2 

120.4 

3.2 

-176.9 

2c 

1.248 

1.388 

1.437 

120.0 

120.3 

-3.2 

-176.6 

3a 

1.248 

1.384 

1.436 

121.4 

122.2 

8.1 

-173.1 

3b 

1.248 

1.385 

1.436 

121.2 

122.1 

8.0 

-173.2 

3c 

1.248 

1.385 

1.436 

121.3 

122.2 

8.1 

-173.1 

4a 

1.248 

1.384 

1.437 

121.3 

122.3 

7.5 

-173.5 

4b 

1.247 

1.380 

1.437 

121.3 

122.8 

7.0 

-174.9 

4c 

1.249 

1.384 

1.439 

121.4 

121.5 

11.6 

-173.5 

5a 

1.247 

1.396 

1.450 

120.5 

119.7 

-10.7 

173.0 

5b 

1.247 

1.391 

1.494 

120.6 

119.7 

-10.4 

173.3 

5c 

1.248 

1.391 

1.492 

120.2 

119.7 

-12.5 

171.2 


© 2006 by Taylor & Francis Group, LLC 







Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


207 


TABLE 10.5 


Comparison of Dipole Moments, Mulliken Charges (Electrons), and MEPs 


Compound 

Carbonyl O Atom 

Amide N Atom 

C 7 Atom 
Charge 

Dipole Moment, 
Debye 

Charge 

Mil ' 1 

Charge 

MEP a 

JH-mimic 

-0.39 

-73.2 

-0.32 

-37.3 

0.38 

2.02 

la 

-0.35 

-75.0 

-0.33 

-23.2 

0.34 

3.68 

lb 

-0.37 

-74.5 

-0.33 

-25.1 

0.30 

3.25 

lc 

-0.36 

-75.5 

-0.33 

-30.0 

0.35 

3.55 

2a 

-0.35 

-73.1 

-0.33 

-22.8 

0.35 

3.82 

2b 

-0.36 

-76.2 

-0.34 

-17.0 

0.35 

3.45 

2c 

-0.37 

-75.7 

-0.34 

-18.7 

0.35 

4.37 

3a 

-0.36 

-75.9 

-0.37 

-20.3 

0.34 

3.63 

3b 

-0.37 

-76.7 

-0.37 

-27.8 

0.34 

3.25 

3c 

-0.36 

-77.3 

-0.37 

-22.1 

0.34 

3.74 

4a 

-0.36 

-74.5 

-0.37 

-25.8 

0.34 

3.22 

4b 

-0.37 

-75.4 

-0.38 

-26.7 

0.30 

3.46 

4c 

-0.37 

-75.7 

-0.35 

-38.6 

0.35 

3.55 

5a 

-0.35 

-73.8 

-0.31 

-33.0 

0.34 

3.52 

5b 

-0.35 

-75.6 

-0.31 

-33.1 

0.34 

3.56 

5c 

-0.36 

-75.7 

-0.32 

-30.2 

0.35 

3.27 


a Most negative electrostatic potential located by indicated atom expressed in kcal/mol superimposed on the isodensity 
surface (0.002 e/au 3 ) of the molecule. 



JH-mimic 


FIGURE 10.3 (See color insert following page 204.) Molecular electrostatic potential plotted onto the total electron 
density surface defined as 0.002 e/au 3 (essentially the van der Waals surface) of the maximum energy conformer of JH 
where R = R 1 = methyl, JH-mimic, and deet (compound la) (Table 10.2). The surface is color-coded according to the 
magnitude of its potential in units of kcal/mol. The red region is by the carbonyl oxygen atom. (From A. K. Bhattacharjee, 
R. K. Gupta, D. Ma, and J. M. Karle, Journal of Molecular Recognition , 13, 213, 2000.) 




JHmax 



© 2006 by Taylor & Francis Group, LLC 








208 


Insect Repellents: Principles, Methods, and Uses 


hydrogen atoms in the molecules, in the range of 15.4-35.2 kcal/mol. The calculated charge densities on 
the amide nitrogen atom and the C 7 atom are also found to be quite close between JH-mimic and the deet 
compounds, keeping in mind the diverse nature of the substituents in other parts of the molecules. 

Furthermore, the profiles of electrostatic potential beyond the van der Waals surface at a constant 
potential of — 10.0 kcal/mol are comparable to a large negative potential region localized by the amide . 27 
Electrostatic potential characteristics beyond the van der Waals surface of the molecules are believed to 
be the key features primarily responsible for recognition interaction between an approaching molecule 
and its receptor at longer distances of separation . 32 It is through this potential that a molecule reacts with 
any other system in its vicinity, recognizes its receptor, and accordingly promotes interaction between 
the complimentary sites. 

The electrostatic potential of functional groups that are commonly found in diphenylether and 
terpenoid JH mimics are similar to each other in terms of their electrostatic potential characteristics. 
This electrostatic bioisosterism has led to the understanding of the universality of active structures and 
aided in the design of new active analogs . 52 Therefore, the present investigation indicates that the 
electrostatic characteristics of the deet compounds are likely to cause similar recognition interactions 
with the JH receptor as the JH of the insects at a distance to promote binding interactions. 

It is interesting to note that the localized negative potential region by the amide moiety in the deet 
compounds is qualitatively linked to their potent repellent activity, with the less-potent repellent 
compounds having a more extended, and therefore a more diffuse, negative potential zone . 27 It 
appears that a more localized negative potential region in the amide group, as seen with JH-mimic, is 
consistent with higher protection times. Because the similarity of the negative potential profiles at 
— 10.0 kcal/mol seems to play a role in the repellent potency of deet analogs, this observation should aid 
in the design of potent analogs of this class of insect repellents. 

Large hydrophobic regions in the molecule appear to be necessary for both recognition and potent 
repellent activity. Hydrophobic effects are the result of averaged electrostatic interaction of the molecule 
with its surroundings, solvent, and protein environment. Sites of nonpolar or weakly polar regions in 
different molecules tend to come together to escape contact with water and to minimize the dehydration 
free energies . 45 Thus, matching the nonpolar regions of ligands with the receptor sites gives a reasonable 
measure of hydrophobic complementarity, and also represents the stabilization of the enzyme-substrate 
or ligand-receptor complex. 

Polarity of a certain region in the molecule can be regarded as proportional to the electrostatic field. 
A strong electrostatic field of a molecule attracts molecules having large dipoles, such as water, while the 
weak electrostatic field regions of the molecule do not attract water molecules and are, therefore, 
hydrophobic . 45 ' 46 Different approaches have recently appeared to theoretically represent hydrophobic 
interactions in terms of local solute-solvent electrostatics . 56 However, a simple assessment of 
hydrophobic similarity may be carried out by determining the distribution of charges or electrostatic 
potentials at different regions on the van der Waals surface of the molecule. The observed low-dipole 
moments (< 4 Debye) of JH-mimic and the deet analogs also correspond to the lipophilic nature of the 
compounds. Because olfactory sensations of the insects require some degree of lipid solubility , 36 
hydrophobicity of the repellents is likely to be an important factor for potent repellent activity. 

Correlation of Molecular Orbital Properties in JH-Mimic and in Deet and Its Analogs 

The deet compounds have similar highest occupied molecular orbital (HOMO)-lowest unoccupied 
molecular orbital (LUMO) energy gaps as shown by the rj values in Table 10.6, an index of intrinsic 
reactivity . 57 These values indicate that the deet compounds are nearly similar in intrinsic reactivity. The 
energy of HOMO and LUMO orbitals plays a major role in governing chemical reactions. The energy 
difference between the orbitals is known as the electronic band gap, and is often responsible for the 
formation of many charge-transfer complexes . 57 Table 10.6 shows a relatively constant E H omo with a 
large negative value and a more variable and much smaller magnitude E LUMO , implying a greater role of 
LUMO or electron-acceptor ability of the compounds than their electron donating. Therefore, the 


© 2006 by Taylor & Francis Group, LLC 



Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


209 


TABLE 10.6 

HOMO and LUMO Eigenvalues 


Eigenvalues, eV 


Compound 

HOMO 

LUMO 

rj — (Elljmo ~ Ehomo)/2 

JH-mimic 

-9.847 

0.98 

5.41 

la 

-9.542 

0.146 

4.84 

lb 

-9.555 

1.514 

5.53 

lc 

-9.207 

0.137 

4.67 

2a 

-9.589 

-0.187 

4.70 

2b 

-9.518 

0.027 

4.77 

2c 

-9.256 

0.011 

4.63 

3a 

-9.854 

-0.090 

4.88 

3b 

-9.274 

-0.063 

4.60 

3c 

-9.599 

-0.105 

4.74 

4a 

-9.593 

-0.051 

4.77 

4b 

-9.889 

1.536 

5.71 

4c 

-9.319 

-0.047 

4.63 

5a 

-9.514 

0.111 

4.81 

5b 

-9.495 

0.151 

4.82 

5c 

-9.236 

0.111 

4.67 


electron transfer from a suitable receptor molecular orbital to the LUMO of the deet compounds, rather 
than a donation of electrons from the deet compounds, seems a more plausible mechanism for 
the compounds. 

Molecular Electronic Properties of JH 

Stereoelectronically, the maximum energy conformer has features adjacent to the carbonyl oxygen atom 
most similar to the deet analogs and JH-mimic. The carbonyl oxygen atom of all the JH conformers is 
the most nucleophilic site, being the most negative potential site on the van der Waals surface in the 
molecule. It varies from —66.1 to —71.3 kcal/mol for the maximum energy conformer. The electrostatic 
potential feature beyond the van der Waals surface generated by the carbonyl oxygen atom of the 
maximum energy conformer is most similar to the deet analogs and JH-mimic, as it has the largest 

— 10 kcal/mol potential surface of the three conformers. 27 

Other electronic features of the JH conformers are similar to the deet analogs and JH-mimic. Again, 
the surface of JH has large hydrophobic regions ( light green regions of potentials ranging between 11.7 
and 12.1 kcal/mol. Figure 10.3). The HOMO, LUMO, and reactivity indices range from —9.574 to 

— 9.602 eV, —0.038 to 0.025 eV, and 4.76 to 4.81 kcal/mol, respectively. The dipole moment of the JH 
conformers varies from 1.83 to 4.45 Debye. Thus, clearly there exists an electronic similarity between the 
natural juvenile hormone molecule, the JH-mimic, and the deet analogs that most likely implies an 
electrostatic bioisoterism between all the molecules. 


Development of a New Model for Repellent Research 
Chemical-Feature Based Considerations 

The factors involved in attracting mosquitoes to a host are complex and are not fully understood. 58 
Mosquitoes use, at the very least, visual, thermal, and olfactory stimuli to locate a host. Of these, 
olfactory cues are probably the most important. It has been estimated that 300^100 compounds are 


© 2006 by Taylor & Francis Group, LLC 








210 


Insect Repellents: Principles, Methods, and Uses 


released from a human body as by-products of metabolism and that more than 100 volatile compounds 
can be detected in human breath (see Chapter 4). Of these odors, only a fraction have been isolated and 
fully characterized. Carbon dioxide and lactic acid are the two best-studied mosquito attractants. Carbon 
dioxide, released mainly from breath but also from skin, serves as a long-range airborne attractant and 
can be detected by mosquitoes at distances of up to 40 m. Lactic acid, in combination with carbon 
dioxide, and uric acid are also highly attractive. 

It is also believed that mosquitoes can sense which host is the richest source of cholesterol and 
B vitamins, nutrients that mosquitoes cannot synthesize. Mosquitoes have chemo-receptors on their 
antennae that are stimulated by lactic acid. 59 It is also speculated that the same receptors may be inhibited 
by deet-based insect repellents. 60 

In a continuation of the efforts to design and discover new insect repellents from structure-activity 
relationship studies 27,33 and to better understand the mechanism of insect repellency, the authors have 
developed 61 a three-dimensional chemical function-based pharmacophore model for potent arthropod 
repellent activity to provide a foundation for compound database searches to aid the discovery of new 
repellent candidates. We have utilized 3D QSAR-CATALYST® methodology on a training set of 
eleven known structurally diverse insect repellent compounds, including deet, to develop the model 
whose validity applies to a variety of other arthropod repellents beyond that of the training set. 

Significance and Uniqueness of the Methodology 

Thus far, no attempt has been made to design insect repellents rationalizing the pharmacophores obtained 
from the similarity analysis of studies on stereoelectronic properties. The authors developed a 
pharmacophore from a training set of deet and its eleven analogues using 3D QSAR. This was 
accomplished by utilizing the existing expertise and CATALYST computer software 62 at Walter Reed 
Army Institute of Research (WRAIR), Silver Spring, Maryland, U.S.A. The prerequisite for developing a 
reliable 3D-QSAR model for a novel insect repellent compound is the correlation of a characteristic and 
reproducible biological activity to structural information of the respective compound. The confor¬ 
mational model of the compound in the training set has enabled us to use the best three-dimensional 
arrangement of chemical functions predicting the repellent activity variations among the compounds in 
the training set. The pharmacophore has also facilitated the search for compound databases to identify 
new repellent compounds. 

Computational Methods and Materials 

Procedure for Development of the 3D-QSAR Pharmacophore Model 

The 3D-QSAR study was performed using CATALYST 4.8 software. 62 The algorithm treats molecular 
structures as templates composed of chemical functions localized in space that will bind effectively with 
complementary functions on the respective binding proteins. The most relevant biological features are 
extracted from a small set of compounds that cover a broad range of activity. 63 

This process makes it possible to use structure and activity data for a set of lead compounds to generate 
a pharmacophore representative of the activity of the lead set. At the heart of the software is the HypoGen 
algorithm that allows identification of pharmacophores that are common to the “active” molecules in the 
training set but are absent in the “inactives.” 64 Structures of the arthropod repellent compounds 
(Table 10.7) were edited within CATALYST and energy minimized to the closest local minimum 
using the generalized CHARMM-like forcefield as implemented in the program. Molecular flexibility 
was taken into account by considering each compound as an ensemble of conformers representing 
different accessible areas in a three-dimensional space. The “best searching procedure” was applied to 
select representative conformers within 10 kcal/mol of the global minimum. 65 


* Registered trademark of Accelrys Inc., San Diego, CA. 


© 2006 by Taylor & Francis Group, LLC 



Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


211 


TABLE 10.7 


Names and Activities of the Repellents Used to Create the Training Set 

Compound 

Name 

Repellent Activity (in h of PT a ) 

1 

Deet (A/^Af-diethyl-m-toluamide) 

1.0 

2 

Af,Af-Diethyl-2-ethoxy benzamide 

0.5 

3 

Af,Af-Dipropyl-2-benzyloxyacetate 

0.5 

4 

1 -butyl-4-methylcarbostyril 

2.0 

5 

A/,Af-Dipropyl-2-ethoxybenzamide 

0.3 

6 

2-butyl-2-ethyl-1,3-propanediol 

1.7 

7 

1,3-bisbutoxymethyl-2-imidazol 

0.6 

8 

Af,Af-Diethyl-2-chlorobenzamide 

1.2 

9 

Hexachlorophenol 

0.2 

10 

1,3-propanediolmonobenzoate 

7.5 

11 

Diisobutylmalate 

2.5 


d PT is protection time in hours provided by the repellent compounds. 


Conformational models of the training set of 11 repellents were generated that emphasize 
representative coverage within a range of permissible Boltzmann population with significant abundance 
(within 10.0 kcal/mol) of the calculated global minimum. This conformational model was used for 
pharmacophore generation within CATALYST, which aims to identify the best three-dimensional 
arrangement of chemical functions, such as hydrophobic regions, hydrogen bond donor, hydrogen bond 
acceptor, and positively or negatively ionizable sites, distributed over a three-dimensional space 
explaining the activity variations among the compounds in the training set. The hydrogen bonding 
features are vector functions, whereas all other functions are points. 

Pharmacophore generation was carried out by setting the default parameters in the automatic generation 
procedure in CATALYST (function weight = 0.302, mapping coefficient = 0, resolution = 260 pm, 
andactivity uncertainty = 3). An uncertainty “A” in the CATALYST paradigm indicates an activity 
value lying somewhere in the interval from “activity divided by A” to “activity multiplied by A.” The 
statistical relevance of the obtained pharmacophore is assessed on the basis of the cost relative to the null 
hypothesis and the correlation coefficient. 62 ’ 64 The pharmacophores are then used to estimate the 
activities of the training set. These activities are derived from the best conformation generation model of 
the conformers displaying the smallest root-mean square (RMS) deviations when projected onto the 
pharmacophore. HypoGen considers a pharmacophore to be one that contains features with equal 
weights and tolerances. Each feature (e.g., hydrogen-bond acceptor, hydrogen-bond donor, hydrophobic 
regions, positive ionizable group, etc.) contributes equally to estimate the activity. Similarly, each 
chemical feature in the HypoGen pharmacophore requires a match to a corresponding ligand atom to be 
within the same distance of tolerance. 64 The method has been documented to perform better than a 
structure-based pharmacophore generation. 63 

Bioassay for Mosquito Repellency 

The new arthropod repellent candidates were tested for repellent efficacy against Aedes aegypti using an 
in vitro blood feeding system. The in vitro test system provided an estimate of the amount of repellent 
that must be applied to produce a given level of effectiveness against an arthropod test population 
(i.e., the compound’s inherent repellency). Mosquitoes were reared under standardized conditions and 
held in a cage at 27°C and 75% RH until testing. This test system consisted of a mosquito blood feeder, a 
constant-temperature water circulator, and a specially designed cage. The mosquito blood-feeder 
contained five circular blood reservoirs, each of which was filled with outdated human blood and 
covered with the candidate repellent-treated Baudruche membrane. In the beginning, the candidate 
repellents were diluted in ethanol to provide concentrations of 0.02, 0.04, 0.08 and 0.16 mg/cm 2 . The test 


2006 by Taylor & Francis Group, LLC 



212 


Insect Repellents: Principles, Methods, and Uses 


materials, including the control, were applied randomly to the five separate membrane positions. Then, 
250 female mosquitoes (5-15 days old) were given access to the blood reservoirs on a “free choice” basis 
by sliding back a door in the floor of the test cage. The number of mosquitoes probing and feeding on 
each well was noted at 2-min intervals. The test was terminated at the end of 20 min. The test results were 
expressed as the total of ten feeding counts. The effective dose was then calculated from a probit analysis 
of the feeding count obtained in the respective tests. The statistical distribution of tested chemical 
sensitivity levels for Aedes aegypti was calculated from the dose-response regression equation. 66-68 

Results and Discussion 

The three-dimensional chemical function or feature-based pharmacophore for arthropod repellent 
activity of a compound developed in the present study was found to contain two aliphatic hydrophobic 
functions, one aromatic hydrophobic (aromatic ring) function and one hydrogen bond acceptor function 
in specific geometric locations surrounding the molecular space (Figure 10.4). This implies that an insect 
repellent compound needs to have the physico-chemical characteristics described above to have 
potent activity. 

The pharmacophore model was generated by creating a training set of 11 structurally diverse known 
arthropod repellent compounds having a broad range of repellent activities as shown in Figure 10.5 
(diagrams of the structures found in Table 10.7). The repellent activity of the 11 repellent compounds in 
the training set that includes deet covers a broad range of activity, from an ED 50 of about 1 pg/cm 2 to 
about 50 mg/cm 2 (Table 10.8 and Tablel0.9). CATALYST methodology 62 was used to develop the 
model by placing suitable constraints on the number of available chemical features, such as aromatic 
hydrophobic or aliphatic hydrophobic interactions, hydrogen bond donors, hydrogen bond acceptors, 
hydrogen bond acceptors (lipid), and ring aromatic sites, to describe the arthropod repellent activity of 
the compounds. Earlier reported 27 results of quantum chemical calculations and the stereoelectronic 
properties of these compounds provided guidance for selection of these physico-chemical features. 



FIGURE 10.4 Pharmacophore model for insect repellent activity. It is characterized by two hydrophobic aliphatic 
functions, one aromatic function, and one hydrogen bond acceptor function. The hydrogen bonding feature is a vector; 
whereas, all other functions are points. The sphere indicates the tolerance area under the specific function. (From 
A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and R. K. Gupta, QSAR and Combinatorial Science , 24, 593, 2005.) 


© 2006 by Taylor & Francis Group, LLC 






























Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


213 



FIGURE 10.5 The correlation diagram between the protection time (insect repelling time) conferred by the compounds in 
the trainning set and their predicted protection time (R = 0.9). (From A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and 
R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.) 

During the pharmacophore development, molecules were mapped to the features with pre-determined 
conformations generated using the “fast fit” algorithm in CATALYST. The conformational energy for 
developing the set of three-dimensional conformers ranged between 0 and 20 kcal/mol. The procedure 
resulted in the generation of 10 alternative pharmacophores for repellent activity of the compounds and 

TABLE 10.8 


Predicted and Experimentally Determined Protection Times of 
the Repellents in the Training Set 


Compd. 

Experimental 

PT (h) 

Predicted 

PT (h) 

Error 

1 

1.0 

1.4 

1.4 

2 

0.5 

0.73 

1.5 

3 

0.5 

0.52 

1.0 

4 

2.0 

1.7 

-1.2 

5 

0.3 

0.17 

-1.8 

6 

1.7 

0.68 

-2.5 

7 

0.6 

0.44 

-1.4 

8 

1.2 

2.1 

1.7 

9 

0.2 

0.3 

1.5 

10 

7.5 

14.0 

1.9 

11 

2.5 

2.6 

1.0 


2006 by Taylor & Francis Group, LLC 



214 


Insect Repellents: Principles, Methods, and Uses 


TABLE 10.9 


Description of Cost Analysis of the Pharmacophores 3 


Hypothesis 

Total Cost 

Fixed Cost 

RMS 

Correlation 

1st 

44.9356 

40.0875 

0.82985 

0.918127 

2nd 

45.4925 

40.0875 

0.84233 

0.908573 

3rd 

45.8306 

40.0875 

0.84894 

0.895599 

4th 

46.8375 

40.0875 

0.88776 

0.881202 

5th 

47.6982 

40.0875 

0.89736 

0.879984 

6th 

48.9341 

40.0875 

0.90091 

0.872108 

7th 

49.0084 

40.0875 

0.92756 

0.866847 

8th 

49.8163 

40.0875 

0.91329 

0.852024 

9th 

50.2185 

40.0875 

0.98337 

0.833532 

10 

51.1013 

40.0875 

0.98973 

0.807611 

Null cost 

110.8533 

40.0875 

1.00971 

0.0 


Log output file showing the calculated statistics. 


appeared to perform quite well for the training set. The correlation coefficients ranged from 0.91 to 0.87 
for six of the ten models. The total costs of the pharmacophores varied over a narrow range and the 
difference between the fixed cost and the null cost was 71 bits, satisfying the acceptable range as 
recommended in the cost analysis of the CATALYST procedure. 62 ' 64 

Significantly, the best pharmacophore, characterized by two hydrophobic aliphatic functions, one 
aromatic ring function, and one hydrogen bond acceptor function (Figure 10.4), was also statistically 
the most relevant pharmacophore. The predicted arthropod repellent activity values, along with the 
experimentally determined protection time (in hours) for repellent activity, of the compounds are 
presented in Table 10.2. A plot of the protection time conferred by the compounds in the training set and 
their predicted protection time demonstrated a good correlation (/? = 0.91), indicating the predictive 
power of the pharmacophore (Figure 10.5). The highly potent analogues of the series mapped all the 
functional features of the best hypothesis with high scores (e.g., Compd. 1, Figure 10.6a), whereas the 
less-potent compounds mapped fewer of the features (e.g., Compd. 9, Figure 10.6b). In order to further 
cross-validate the model, it was mapped onto four other earlier studied repellent candidates (Figure 10.7) 
in the authors’ laboratory: (1) A,A-diethyl-2-(3-trifluoromethyl-phenyl)-acetamide (PT = 0.14h), 
(2) 2-cyclohexyl-ACV-diethylacetamide (PT = 0.24h), (3) A,A-diethyl-2-(3-bromo-phenyl)-acetamide 
(PT = 0.63h), and (4) A,A-diethyl-3-trifluromethyl-benzamide (PT = 0.5h). All of these compounds 
map the pharmacophore in varying degrees (Figure 10.7a through d). 

To further examine the validity of the pharmacophore, it was mapped on a compound recently reported 
in the literature, a novel 18-carbon acid, isolated from samples of greasy gaur hair, that was found to 
have insect-repellent activity and can function as a landing and feeding deterrent to mosquitoes (see 
Chapter 3). 69 Surprisingly, the pharmacophore mapped extremely well on this molecule, proving the 
consistency in the predictive power for insect repellent activity of the model (Figure 10.7e) 61 

The pharmacophore allowed the authors to screen the in-house WRAIR-Chemical Information System 
(WRAIR-CIS) database 70 to search for candidate arthropod repellent compounds. The WRAIR-CIS 
database has over 290,000 compounds; it was transformed into a multi-conformer database in 
CATALYST using the catDB utility program as implemented in the software. 62 The catDB 
format allows a molecule to be represented by a limited set of conformations, thereby permitting 
conformational flexibility to be included during the search of the database. The authors have utilized the 
best fit mapping of the pharmacophore in potent analogues by using a fast-fit algorithm, a principle 
component analysis, a partial least squares technique, a linear regression technique, or a non-linear 
regression technique. 


© 2006 by Taylor & Francis Group, LLC 





Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


215 




FIGURE 10.6 (See color insert following page 204.) Pharmacophore mapping onto (a) Compd. 1, deet (a highly potent 
repellent) and (b) Compd. 9 (a less potent repellent). (From A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and 
R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.) 


© 2006 by Taylor & Francis Group, LLC 














216 


Insect Repellents: Principles, Methods, and Uses 




FIGURE 10.7 (See color insert following page 204.) Pharmacophore mapping on other known repellents showing the 
cross-validation of the pharmacophore. Protection time of these agents are shown in the parentheses: (a) N,Af-diethyl-2- 
(3-trifluoromethyl-phenyl)-acetamide (PT = 0.14 h), (b) 2-cyclohexyl-iV,Af-diethylacetamide (PT = 0.24 h), (c) N,N- 
diethyl-2-(3-bromo-phenyl)-acetamide (PT = 0.63 h), (d) A/,A/'-diethyl-3-trifluoromethyl-benzamide (PT = 0.5 h), and 
(e) onto 5-[5-(l-hydroxy-nonyl)-tetrahydro-furan-2-yl]-pentanoic acid. (From A. K. Bhattacharjee, W. Dheranetra, D. A. 
Nichols, and R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.) 


© 2006 by Taylor & Francis Group, LLC 

















218 


Insect Repellents: Principles, Methods, and Uses 



FIGURE 10.7 Continued. 


The pharmacophore was finally converted into a 3D shape-based template containing all the chemical 
features necessary for potent arthropod repellent activity and used for WRAIR-CIS database searches. 
After each compound in the WRAIR-CIS was converted into 3D multi-conformations, with an energy 
range of 0-20 kcal/mol using the catDB algorithm of CATALYST, the full data set was stored in an SGI 
Octane workstation. The result of the search led us to identify 138 compounds for repellent activity. The 
down selection of the identified compounds was carried out by evaluating the in silico ADME/Toxicity 
properties and choosing only those compounds that had favorable properties. ADME/Toxicity 
evaluations were carried out by using Cerius 2 and TOPKAT methodology, 71 ' 72 as implemented in 
these software applications. 

The overall procedure of compound identification and selection was carried out in an iterative manner 
by generating several shape-based pharmacophore templates on a few potent repellent compounds. 
Ultimately, it was possible to shortlist four compounds (Figure 10.8 and Figure 10.1 1) that were found to 
exhibit remarkable repellent activity, fulfilling the important goals of an ideal repellent. One of these four 
compounds, 2-methyl-l-(2,3,5,6-tetramethyl-phenyl)-propan-l-one appears to fulfill most of the goals 
for developing an ideal repellent. The four compounds are presented with the protection time in the 
parentheses of each of them: (1) 2-bromo-l-(2,5-dimethoxy-phenyl)-ethanone (PT = 2.6h); 
(2) 2-methyl-l-(2,3,5,6-tetramethyl-phenyl)-propan-l-one (PT = 9.3h); (3) 2-allylsulfanyl-3-methyl- 
pyrazine (PT=1.6h); and (4) 2-(2-chloro-phenoxy)-2-methyl-propionamide (PT = 2.65h). Mappings 
of the pharmacophore on these four compounds are shown in Figure 10.8a through d. Although all of the 
compounds possess outstanding in vitro arthropod repellent activity and have reasonably well-tolerated 
properties for promising repellent candidates, the compounds have yet to be tested for in vivo efficacy 
and toxicity. 

Thus, the pharmacophore model for repellent activity allowed the authors to successfully search for 
compounds in databases and identify four lead repellent candidates that are currently under further 
investigation. Although no model is perfect, regardless of what it represents, virtual screening of 


© 2006 by Taylor & Francis Group, LLC 




Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


219 



(b) 



FIGURE 10.8 (See color insert following page 204.) Pharmacophore mapped onto four new insect repellents discovered 
through database searches by using the pharmacophore: (a) 2-bromo-l-(2,5-dimethyoxy-phenyl)-ethanone, (b) onto 2- 
methyl-l-(2,3,5,6-tetrahmethyl-phenyl)-propan-l-one, (c) onto 2-allylsufanyl-3-methyl-pyrazine, and (d) onto 2-(2-chloro- 
phenoxy)-2-methyl-propionamide. (From A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and R. K. Gupta, QSAR and 
Combinatorial Science, 24, 593, 2005.) 


© 2006 by Taylor & Francis Group, LLC 

























































220 


Insect Repellents: Principles , Methods, and Uses 




(«0 

FIGURE 10.8 Continued. 


© 2006 by Taylor & Francis Group, LLC 


Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


221 




FIGURE 10.9 Structure of compounds in the training set. (From A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and 
R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.) 


hundreds of compounds from databases to identify potential hits in a relatively short period of time has 
opened a new dimension for the search for new arthropod repellent candidates. Even though it is a 
complex, expensive, and time-consuming path to develop a perfect repellent from the discovery stage to 
the shelf for over-the-counter sale, scientists are continuing to explore new strategies to efficiently 
minimize the amount of effort required and to translate effectively the potent intrinsic physico-chemical 
characteristics of the compounds into new candidates with superior properties for suitable skin 
application to protect against arthropod biting. 


Concluding Remarks and Future Perspectives 

The stereoelectronic properties, similarity analysis, and the 3D pharmacophore models in the above 
studies could satisfactorily explain the insect repellent properties of the compounds. The pharmacophore 
model made it possible to search compound databases to identify new repellent candidates. 

The first investigation on the electronic properties of 31 repellents suggests that the properties 
of the amide group (N-C=0 atoms) in these compounds play a key role in determining the duration of 
the protection against mosquito bites. The substituents attached to carbon and nitrogen atoms of the amide 
group together influence the electronic properties of the amide group. Thus, a balance of polarity between 
the two parts of the molecule seems to be an important contributing factor for potent repellent activity. 


2006 by Taylor & Francis Group, LLC 



222 


Insect Repellents: Principles , Methods, and Uses 



FIGURE 10.9 Continued. 

The investigation of the comparison of stereoelectronic properties of deet compounds with JH-mimic 
and JH unraveled a few important facts about these compounds at a molecular level, not only providing a 
better insight into the mechanism of action of the deet repellents, but also facilitating the design of more 
efficacious deet-like compounds. The results of the study indicate a model for similar molecular 
recognition of the deet compounds and the JH-mimic where the three crucial factors appear to be: 

• Considerable steric similarity between the amide moiety of the compounds 

• Similarity of electrostatic properties and profiles beyond the van der Waals surface 

• Similarity of a large distribution of a weak electrostatic field on the van der Waals surface 


© 2006 by Taylor & Francis Group, LLC 



Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


223 



FIGURE 10.9 Continued. 

Although electrostatic similarity beyond the van der Waals surface is considered to be the primary 
index for the molecular recognition between compounds, the similarity of the steric components, 
Mulliken charges, and negative potential adjacent to the oxygen and nitrogen atoms of the amide moiety 
may all contribute significantly to the overall mechanism of repellent action of the deet compounds. On a 
molecular level, the repellent action of the deet compounds may be attributed to avoiding a host-guest 
complementarity conflict with the receptor. 

The stereoelectronic property study provides three important guidelines to effectively design this class 
of insect repellents. Specifically, there needs to be: 

• An amide moiety on one end of the molecule that contains a charge separation between the 
oxygen and nitrogen atoms to facilitate a strong electronic interaction with the receptor 

• Electrostatic similarity to JH or its mimic molecules 

• A large, weakly charged region to facilitate optimum hydrophobic interaction with the receptor 
that may be a long-chain hydrocarbon and need not be an aromatic ring 

Thus, the study has illustrated: (1) the electrostatic bioisosterism of juvenile hormone, its mimic, and 
the repellents; (2) a probable mode of action of insect repellent activity; and (3) additional 
stereoelectronic features that may be added to the previous report on the design of insect repellents. 
However, possible pitfalls for designing potent repellents based on these criteria may not be ruled out if 
the designed molecules are either too volatile or insufficiently volatile with R groups too bulky for fitting 
into the receptor site. 

The 3D-QSAR pharmacophore study on repellents demonstrated a new computational approach for 
organizing the molecular characteristics of a set of structurally diverse arthropod repellents to a model 
that may be both statistically and mechanistically significant for potent repellent activity and may have 
applicability beyond the bounds of known repellents. The resulting model can also be used to unravel a 
possible rationale for the target-specific arthropod repellent activity of these compounds. The chemically 
significant molecular characteristics disposed on a three-dimensional space generated a pharmacophore 
that is found to be quite satisfactory in correlating experimental repellent activity with the predicted 


2006 by Taylor & Francis Group, LLC 



224 


Insect Repellents: Principles, Methods, and Uses 



/V,A/-Diethyl-2-(3-trifluoromethyl-phenyl)-acetamide 
(PT = 0.14 hr) 



2-Cyclohexyl-A/,A/-diethylacetamide 
(PT = 0.24 hr) 



/V,/V-Diethyl-2-(3-bromo-phenyl)-acetamide 
(PT = 0.63 hr) 



N,N-Diethyl-3-trifluromethyl-benzamide 
(PT = 0.5 hr) 



5-[5-(1 -Hydroxynonyl)tetrahydrofuran-2-yl]pentanoic acid 

FIGURE 10.10 Structure of known insect repellents used for validating the pharmacophore. (From A. K. Bhattacharjee, 
W. Dheranetra, D. A. Nichols, and R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.) 



2-Bromo-2\5'-dimethoxyacetophenone 

(PT=2.6hr) 



2-Allylsulfanyl-3-methyl-pyrazine 

(PT=1.6hr) 



2-Methyl-1 -(2,3,5,6-tetramethyl 
phenyl)propan-1-one 


2-(2-Chlorophenoxy)-2-methyl 

propionamide 


(PT=9.3hr) 


(PT=2.7hr) 


FIGURE 10.11 Structure of four new insect repellents discovered using the pharmacophore. (From A. K. Bhattacharjee, 
W. Dheranetra, D. A. Nichols, and R. K. Gupta, QSAR and Combinatorial Science, 24, 593, 2005.) 


© 2006 by Taylor & Francis Group, LLC 



Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


225 


activity of the compounds (R = 0.9). Potent repellent activity appears to be favored by two aliphatic 
hydrophobic functions, one aromatic hydrophobic function (aromatic ring) and one hydrogen bond 
acceptor function in specific geometric locations surrounding the molecular space. 

The validity of the pharmacophore, which extends to structurally different classes of compounds, 
allowed us to discover new repellent candidates, and thereby provides a powerful template for 
identification of novel arthropod repellent candidates. Because the identity of the biological target 
for arthropod repellent activity remains unknown, this 3D-QSAR pharmacophore should aid in the 
design of well-tolerated, target-specific arthropod repellent active ingredients. The success of discovery 
of new repellent candidates in this study suggests that the 3D-QSAR studies on repellents cannot only 
facilitate the examination of databases to identify new candidates, but also could be a great benefit in 
synthetic efforts to discover better repellents for practical use. 

Although the process of arthropod repellent discovery and development is a long and continuous 
endeavor, in silico technologies can undoubtedly help in reducing the rapidly increasing costs of 
developing new active ingredients. Molecular modeling techniques using in silico tools are uniquely 
suitable for integrating new knowledge about molecular structure with new knowledge about 
repellent activity. 

Acknowledgments 

Material has been reviewed by the Walter Reed Army Institute of Research. There is no objection to its 
presentation and/or publication. The opinions or assertions contained herein are the private views of the 
author, and are not to be construed as official, or as reflecting true views of the Department of the Army or 
the Department of Defense. Research was conducted in compliance with the Animal Welfare Act and 
other federal statutes and regulations relating to animals and experiments involving animals and adheres 
to principles stated in the Guide for the Care and Use of Laboratory Attimals, NRC Publication, 
1996 edition. 

We also wish to express our profound thanks to Ms. Linette Sparacino, from ANTEON, Ft. Detrick, 
MD 21702-5012, for reading and providing invaluable suggestions for improving the manuscript. 

References 

1. P. Buchwald and N. Bodor, Computer-aided drug design: The role of quantitative structure-property, 
structure-activity and structure-metabolism relationships (QSPR, QSAR, QSMR), Drug Future, 27, 
577, 2002. 

2. D. Janseen. The power of prediction. Drug Discovery and Development, 5(1), 38, 2002. 

3. B. L. Podlogar. I. Muegge, and L. J. Brice, Computational methods to estimate drug development 
parameters, Current Opinion in Drug Discovery and Development, 12, 102, 2001. 

4. O. Guner (Ed.), Pharmacophore Perception, Development and Use in Drug Design, La Jolla, CA: 
International University Line Biotechnology Series, 2000. 

5. R. K. D. Peterson, Insects, disease, and military history, American Entomologist, 41, 147, 1995. 

6. R. F. Harwood and M. T. James (Eds.), Entomology in human and animal health, 7th Ed., New York: 
MacMillan, 1979. 

7. B. F. Eldridge and J. E. Edman (Eds.), Medical Entomology: A Textbook on Public Health and 
Veterinary Problems Caused by Arthropods, Boston: Kluwer Academic Publishers, 2000. 

8. A. N. Clements, Development, nutrition and reproduction, The Biology of Mosquitoes. Vol. 1. London: 
Chapman and Hall, 1992. 

9. J. Goddard, Physician’s Guide to Arthropods of Medical Importance, 2nd Ed.. Boca Raton, FL: CRC 
Press, 1996. 

10. M. W. Service, Medical Entomology for Students, London: Chapman and Hall, 1996. 

11. M. Bates, The Natural History of Mosquitoes, New York: MacMillan, 1949. 

12. S. E. Christophers, Aedes Aegypti (L.) The Yellow Fever Mosquito: Its Life History, Bionomics, and 
Structure, London: Cambridge University Press, 1960. 


2006 by Taylor & Francis Group, LLC 



226 


Insect Repellents: Principles, Methods, and Uses 


13. A. N. Clements, The Physiology of Mosquitoes, Oxford: Pergamon Press, 1963. 

14. J. D. Gillett, Mosquitoes, London: Weidendield and Nicolson, 1971. 

15. J. D. Gillett, The Mosquito: Its Life, Activities, and Impact on Human Affairs, Garden City, NY: 
Doubleday, 1972. 

16. W. R. Horsfall, Mosquitoes: Their Behavior and Relation to Disease, New York: Ronald Press, 1955. 

17. S. C. Oaks, V. S. Mitchell, G. W. Pearson, and C. C. J. Carpenter (Eds.), Malaria: Obstacles and 
Opportunities, Washington, DC: Institute of Medicine, National Academy Press, 1991. 

18. World Health Organization, World malaria situation in 1992, Weekly Epidemiological Record, 69, 
309, 1994. 

19. R. K. Gupta et al.. Personal protection measures against arthropods, in Textbook of Military Medicine, 
Vol. 1, D. E. Lounsbury (Ed.), Washington, D.C.: Borden Institute, Walter Reed Army Medical 
Center, 2003, p. 503. 

20. E. T. McCabe, W. F. Barthel, S. I. Gertler, and S. A. Hall, Insect repellents. III. A,A-diethylamides, 
Journal Organic Chemistry, 19, 493, 1954. 

21. I. H. Gilbert. H. K. Gouck, and C. N. Smith, New mosquito repellents, Journal of Economic 
Entomology, 48, 741, 1955. 

22. R. K. Gupta and L. C. Rutledge, Role of repellents in vector control and disease prevention, American 
Journal of Tropical Medicine and Hygiene, 50(6), 82, 1994. 

23. M. S. Fradin, Mosquitoes and mosquito repellents: A clinician's guide. Annals of Internal Medicine, 
128, 931, 1998. 

24. L. A. Thomas. Distribution of the virus of western equine encephalomyelitis in the mosquito vector, 
Culex tarsalis, American Journal of Hygiene, 78, 150, 1963. 

25. A. Nakayama, H. Iwamura, A. Niwa, Y. Nakagawa, and T. Fujita, Development of insect juvenile 
hormone active oxime o-ethers and carbamates. Journal of Agricultural and Food Chemistry, 33, 
1034. 1985. 

26. A. Nakayama and W. G. Richards, A quantum chemical study of insect juvenile hormone mimics: The 
active conformation and the electrostatic similarities. Quantitative Structure-Activity Relationships, 6, 
153, 1987. 

27. A. K. Bhattacharjee, R. K. Gupta, D. Ma. and I. M. Karle, Molecular similarity analysis between insect 
juvenile hormone and _/V,A-diethyl-m-toluamide (deet) analogs may aid design of novel insect 
repellents. Journal of Molecular Recognition, 13, 213, 2000. 

28. J. Boeckh, H. Breer, M. Geier, F. P. Hoever, and B. W. Kruger, Acylated 1,3-aminopropanola as 
repellents against bloodsucking arthropods, Pesticide Science, 48, 359, 1996. 

29. D. Mackay, J. Hubbarde, and E. Webster, The role of QSARs and fate models in chemical hazard and 
risk assessment, in Encyclopedia of Agrochemicals, J. Plimmer (Ed.), New York: Wiley-Interscience, 
2003. 

30. M. V. S. Suryanarayana, K. S. Pandey, S. Prakash, C. D. Raghuveeran, R. S. Dangi, R. V. Swamy, and 
K. M. Rao, Structure-activity relationship studies with mosquito repellent amides. Journal of 
Pharmacological Sciences, 80, 1055, 1991. 

31. R. K. Sharma, S. K. Iain, S. Kumar, and K. M. Rao. Evaluation of some insect repellent formulations. 
Part I—Water soluble ointment bases, Indian Journal Hospital Pharmacy, 21, 26. 1984. 

32. J. S. Murray, B. A. Zilles, K. Jayasuriya, and P. Politzer, Comparative analysis of the electrostatic 
potentials of dibenzofuran and some dibenzo-p-dioxins, Journal of the American Chemical Society, 
108, 915, 1986. 

33. D. Ma, K. Bhattacharjee, R. K. Gupta, and J. M. Karle, Predicting mosquito repellent potency of 
AL/V-diethyl-m-toluamide (deet) analogs from molecular electronic properties, American Journal 
of Tropical Medicine and Hygiene, 60, 1, 1999. 

34. K. I. Judy, D. A. Schooley, L. L. Dunham, M. S. Hall, and B. I. Bergot, Isolation, structure, and 
absolute configuration of new natural insect juvenile hormone from Manduca sexta. Proceedings 
of the National Academy of Sciences, U.S.A., 70, 1509, 1973. 

35. B. J. Bergot, G. C. Jamieson, M. A. Ratcliff, and D. A. Schooley, JH zero: New naturally 
occurring insect juvenile hormone from developing embryos of the tobacco hornworm, Science, 
210, 336, 1980. 


© 2006 by Taylor & Francis Group, LLC 



Discovery and Design of New Repellents by Computer-Aided Molecular Modeling 


227 


36. W. A. Skinner and H. L. Johnson. The design of insect repellents. Drug Design, 10, 277, 1980. 

37. H. L. Johnson et al.. Repellent activity and physical properties of ring-substituted N,N- 
diethylbenzamides. Journal of Economic Entomology, 60, 173, 1967. 

38. S. B. Mclver, A model for the mechanism of action of the repellent deet on Aedes aegypti 
(Diptera: Culicidae), Journal of Medical Entomology , 18, 357, 1981. 

39. S. S. Rao and K. M. Rao, Insect repellent AL/V-diethylphenylacetamide: An update. Journal of 
Medical Entomology, 28, 303, 1991. 

40. E. Fischer et al., Einfluss der configuration auf die Wirkung der Enzyme, Berichte Deutsch 
Chemistry Gesellschaft, 27, 2985, 1894. 

41. H. Weinstein, M. N. Liebman, and C. A. Venazi, in New Methods in Drugs Research, 
A. Makriayannis (Ed.), Vol. I, Barcelona: Prous Science, 1985, p. 233. 

42. A. K. Bhattacharjee, D. Majumdar, and S. Guha, Theoretical studies on the conformational 
properties and pharmacophoric pattern of several bipyridine cardiotonics, Journal of the Chemical 
Society, Perkin Transactions, 2, 805, 1992. 

43. A. K. Bhattacharjee and J. M. Karle, Molecular electronic properties of a series of 4-quinolinecarbi- 
nolamines define antimalarial activity profile. Journal of Medicinal Chemistry, 39, 4622, 1996. 

44. J. Mestres, D. C. Rohrer, and G. M. Maggiora, MIMIC: A molecular-field matching program: exploiting 
applicability of molecular similarity approaches, Journal of Computational Chemistry, 18, 934, 1997. 

45. G. Naray-Szabo and T. Balogh, Viewpoint 7—the average molecular electrostatic field as a QSAR 
descriptor. Part 4. Hydrophobicity scales for amino acid residues, Journal of Molecular Structure: 
THEOCHEM, 284, 243, 1993. 

46. G. Naray-Szabo and T. Balogh, Application of the average molecular electrostatic field in quantitative 
structure-activity relationships, Croatica Chemica Acta, 66, 129, 1993. 

47. A. R. Leach, Molecular Modelling, Principles and Applications, Essex: A.W. Longman Ltd, 566,1998. 

48. M. Randic and G. Krilov, On characterization of molecular surfaces, International Journal of Quantum 
Chemistry, 65, 1065, 1997. 

49. SPARTAN SGI Version 5.1.2., SPARTAN user’s manual, version 5.0. Wavefunction, Inc., Irvine, CA, 
1998. 

50. P. G. Mezey, Shape in Chemistry: An Introduction to Molecular Shape and Topology, New York: VCH, 
1993. 

51. J. D. Petke, Cumulative and discrete similarity analysis of electrostatic potentials and fields, Journal of 
Computational Chemistry, 14, 928, 1993. 

52. A. M. Richard, Quantitative comparison of molecular electrostatic potentials for structure-activity 
studies, Journal of Computational Chemistry, 12, 959, 1991. 

53. W. G. Richards, Quantum Pharmacology, 2nd (Ed.), London: Butterworth, 1983. 

54. A. C. Good, S. S. So, and W. G. Richards, Structure-activity relationships from molecular similarity 
matrices. Journal of Medicinal Chemistry, 36, 433, 1993. 

55. J. Avery, A model for biological specificity, International Journal of Quantum Chemistry, 26,843,1984. 

56. R. G. A. Bone and H. O. Villar, Discriminating D1 and D2 agonists with a hydrophobic similarity index, 
Journal of Molecular Graphing, 13,201, 1995. 

57. R. G. Pearson, The principle of maximum hardness. Accounts of Chemical Research, 26, 250, 1993. 

58. G. R. Bock and G. Cardew (Eds.), Olfaction in Mosquito-Host Interactions, New York: Wiley, 1996. 

59. E. E. Davis and M. F. Bowen, Sensory physiological basis for attraction in mosquitoes. Journal of the 
American Mosquito Control Association, 10, 316, 1994. 

60. E. E. Davis, Insect repellents: Concepts of their mode of action relative to potential sensory mechanisms 
in mosquitoes (Diptera: Culicidae), Journal of Medical Entomology, 22, 237, 1985. 

61. A. K. Bhattacharjee, W. Dheranetra, D. A. Nichols, and R. K. Gupta, 3D pharmacophore model for insect 
repellent activity and discovery of new repellent candidates, QSAR and Combinatorial Science, 24,593, 
2005. 

62. CATALYST Version 4.8 software, San Diego: Accelrys, Inc., 2003. 

63. M. Grigorov, J. Weber, J. M. J. Tronchet, C. W. Jefford, W. K. Milhous, and D. Marie, A QSAR study of 
the antimalarial activity of some synthetic 1,2,4-trioxanes, Journal of Chemical Information and 
Computer Science, 35, 285, 1995. 


2006 by Taylor & Francis Group, LLC 



228 Insect Repellents: Principles, Methods, and Uses 

64. O. A. Gunner, in Pharmacophore, Perception, Development, and Use in Drug Design, University 
International Line (Ed.), San Diego: International University Line, 2000, pp. 17-20. 

65. P. A. Greenridge and J. Weiser, A comparison of methods for pharmacophore generation with the 
catalyst software and their use for 3D-QSAR: Application to a set of 4-aminopyridine thrombin 
inhibitors, Mini-Reviews in Medicinal Chemistry, 1, 79. 2001. 

66. M. Debboun. D. Strickman, T. A. Klein, J. A. Glass, E. Wylie, A. Laughinghouse, R. A. Wirtz, and 
R. K. Gupta, Laboratory evaluation of AI3-37220, AI3-35765, CIC-4, and deet repellents against three 
species of mosquitoes. Journal of the American Mosquito Control Association, 15, 342, 1999. 

67. R. K. Gupta and L. C. Rutledge, Controlled-release repellent formulations on human volunteers under 
three climatic regimens. Journal of the American Mosquito Control Association, 7, 490, 1991. 

68. A. Goldstein, Biostatistics: An Introductory Text, New York: MacMillan, 1964. 

69. J. E. Oliver and K. S. Patterson, Wild ox bugs mosquitoes, Chemical and Engineering News, 49, 2003. 

70. The Chemical Information System, Division of Experimental Therapeutics, Walter Reed Army Institute 
of Research, 503 Robert Grant Avenue, Silver Spring, MD 20910-7500, U.S.A. 

71. Cerius 2 4.9, San Diego: Accelrys, Inc., 2003. 

72. TOPKAT 6.1, San Diego: Accelrys, Inc., 2003. 


© 2006 by Taylor & Francis Group, LLC 



11 


Molecular-Based Chemical Prospecting of Mosquito 
Attractants and Repellents 


Walter S. Leal 


CONTENTS 

Molecular Basis of Insect Olfaction.229 

Choosing Functional Molecular Targets.234 

Screening Techniques.235 

Receptor-Based Approach.235 

Binding Assay-Based Approach.236 

Validating Molecular Targets.237 

Concluding Remarks.239 

Acknowledgments.239 

References.240 


Molecular Basis of Insect Olfaction 

Most insects are primarily reliant on chemical communication to guide their essential behaviors. In 
natural settings, female mosquitoes undoubtedly use airborne chemical signals (semiochemicals) 
integrated with other sensory modalities to find and determine the suitability of hosts for blood 
feeding, sites for oviposition, etc. Female moths, on the other hand, advertise their readiness to mate 
and reproduce by releasing sex pheromones, which are utilized by male moths in odorant-mediated 
navigation toward females. Reception of the semiochemicals by specialized structures in the periphery, 
such as antennae and maxillary palps, is a sine qua non step prior to integration with other stimulus 
modalities in the brain and subsequent translation into behavior. Insect communication, be it host-finding 
in mosquitoes or mate-finding in moths, is a feasible target to disrupt important behaviors. While 
chemical communication-based strategies for monitoring and controlling populations of insects of 
agricultural importance have been extensively used in integrated pest management (IPM) programs, 
similar approaches for reducing mosquito populations and contact between disease vector and host have 
not been fully exploited. 

In general, chemical communication is achieved with airborne semiochemicals that are typically 
available at very low concentrations and buried in complex mixtures of physiologically irrelevant 
compounds. To cope with this, the olfactory system in insects evolved to be highly selective and 


229 


© 2006 by Taylor & Francis Group, LLC 
















230 


Insect Repellents: Principles, Methods, and Uses 


sensitive. To find a semiochemical source; such as blood meal, mate, oviposition site, etc.; insects take 
odorant-mediated flights, which also require a dynamic process for odorant detection. While flying en 
route to a source, insects encounter pockets of semiochemicals separated by clean air spaces. They have 
only a few milliseconds to reset the olfactory system while navigating through clean air. 1 Three major 
groups of proteins play pivotal roles in the dynamics, selectivity, and sensitivity of the insect olfactory 
system. 2 ' 3 They are the odorant receptors (ORs), odorant-binding protein (OBPs), and odorant-degrading 
enzymes (ODEs), which are feasible molecular targets for the development of mosquito attractants and 
novel strategies to reduce mosquito bites. 

Semiochemicals reach the aqueous sensillar lymph through pore tubules (Figure 11.1), but relative 
solubility prevents these hydrophobic molecules from reaching the membrane-bound ORs. The 
semiochemicals are hydrophobic and the sensillar lymph is an aqueous barrier. 2 " 3 Biochemical and 
structural evidence suggest that OBPs selectively bind the physiologically relevant chemical 
compounds 2 ' 4 5 and solubilize them in the form of a ligand-protein complex. 6 While encapsulated by 



FIGURE 11.1 (See color insert following page 204.) Schematic representation of a proposed model for perireceptor 
events in insect olfaction. Odorants enter the sensillar lymph through pore tubules in the cuticle (sensillar wall), are 
solubilized upon being encapsulated by odorant-binding proteins (OBP), and transported to the olfactory receptors. Bound 
pheromone molecules are protected from odorant-degrading enzymes (ODE). Upon interaction with negatively-charged sites 
on the dendritic membrane, the OBP-ligand complex undergoes a conformational change that leads to the ejection of 
pheromone. In BmorPBP. this is achieved by the formation of a C-terminal ot-helix in BmorPBP A that blocks the cavity that 
serves as the binding site in BmorPBP B . After releasing the odorant, the C-terminus may remain in the cavity of BmorPBP 15 
until another odorant is picked up. 56 Note that in this model, the pheromone molecule (not the complex) activates the odorant 
receptor. The signal is terminated by chemical inactivation of the odorant by an odorant-degrading enzyme (ODE). 


© 2006 by Taylor & Francis Group, LLC 












Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents 


231 


OBPs, semiochemicals are not only soluble, but also protected from aggressive ODEs. Then, OBPs carry 
the odorants through the sensillar lymph to the ORs (Figure 11.1). Interaction with negatively charged 
sites on the dendritic membranes 4,7 ' 8 leads to a unique intramolecular rearrangement of the ligand- 
protein complex 9 (Figure 11.1) resulting in the release of odorants. Stopped-flow fluorescence 
measurements show that uptake of pheromones is a rapid process (in the timescale of milliseconds). 
The release of the ligand would be a slow process (half life on the order of 100 s) if not for the 
pH-mediated conformational change that speeds up the delivery of odorants by 10,000-fold. 10 

Although biochemical and structural biology indicate that mosquito OBPs may not undergo the same 
type of intramolecular rearrangement as moth OBPs, the delivery of odorants to the receptors is also 
mediated by a pH-dependent conformational change with a different molecular mechanism. As 
suggested by the crystal structure of an OBP from the malaria mosquito. Anopheles gambiae, 
AgamOBPl, 11 mosquito OBPs possess an overall fold of six helices connected by loops and knitted 
together by three disulfide bridges (Figure 11.2; see also the cover of this book). Although the C-terminus 
of AgamOBPl is too short to form a helix that would occupy the binding pocket at low pH as in the 
silkworm moth’s OBP, BmorPBP, 9 it does form a wall of the binding pocket (Figure 11.2). The 
C-terminus wall is held in place by acid labile hydrogen bonding involving the surrounding helices and 
the N-terminus. The C-terminal carboxylate of valine, Val-125, are within hydrogen bonding distance of 
the hydroxyl of tyrosine, Tyr-54, and of the 8 nitrogen of histidine, His-23. In addition, there are three 
aspartic acid residues, Asp-7, Asp-42, and Asp-118, that interact with either arginine, Arg-5 and Arg-6, 
histidine, His-121, or the backbone nitrogen of Tyr-10. These interactions are likely acid-labile and 
would be disrupted at lower pH, causing both the C- and N-termini to separate. 11 Unlike the formation of 
a C-terminus helix that fits like a piston in the binding pocket of BmorPBP, 9 the C-terminus of 
AgamOPB 1 might move away from the binding pocket. 11 This pH-mediated “unbuckling of the seat 
belt” would expose the ligand to the solvent and, consequently, lower binding affinity at low pH. 

OBPs contribute to the sensitivity of the olfactory system by increasing the capture of molecules 
reaching the sensillar lymph. More importantly, OBPs participate in the selectivity of the olfactory 
system as the conduit between the external environment and the receptors. The remarkable selectivity of 
insect olfactory system is likely to be achieved by “layers of filters,” i.e., by the participation of 
compartmentalized OBPs and olfactory receptors. 3 As suggested by binding assays, 4 ' 5 OBPs transport 
only a subset of compounds that reach the pore tubules. The odorant receptors, on the other hand, are 
activated also by a subset of compounds, as indicated by studies in Drosophila , showing that a single OR 



FIGURE 11.2 (See color insert following page 204.) Three-dimensional structure of AgamOBPl. In the left, three 
disulfide bridges that knit together the scaffold of a-helices are highlighted in cyan. In the right, the acid-labile hydrogen 
bonding involving the C-terminal carboxylate of Val-125 with the hydroxyl of Tyr-54 and the 5 nitrogen of His-23 are shown 
in blue. Although the C-terminus (green) is flexible, Tyr-54 and His-23 are in rigid positions as part of a helices held in place 
by disulfide bridges. 


© 2006 by Taylor & Francis Group, LLC 



232 


Insect Repellents: Principles, Methods, and Uses 


is fired in response to multiple compounds. 12 Even if neither OBPs nor ORs are extremely specific, the 
detection of semiochemicals at the periphery (antennae or maxillary palps) can show remarkable 
selectivity if they function as a two-step filter with only one or very few common ligands. It is worth 
mentioning that OBPs may seem to be selectively less stringent when tested in non competitive binding 
assays. Under physiological conditions, however, OBPs encounter complex mixtures of compounds 
bombarding the sensilla. Therefore, filtering by OBPs may be achieved by selective binding to the key 
stimulus or kinetic competition. 

Semiochemical-OBP interactions are better understood in moths than in mosquitoes. The main 
pheromone-binding protein (PBP is an OBP involved in the reception of a pheromone) from the wild 
silkmoth, Antheraea polyphemus, ApolPBPl, shows apparent high affinity to all three constituents of the 
female-produced sex pheromones: (E,Z)-6,11-hexadecadienyl acetate (E6,Z1 l-16Ac), (E,Z)-6,11- 
hexadecadienal (E6,Z1 l-16Ald), and (E,Z)-4,9-tetradecadienyl acetate (E4,Z9-14Ac). However, 
ApolPBPl shows considerable preference for the major constituent, E6,Z1 l-16Ac, shows lower affinity 
for the shorter acetate, E4,Z9-14Ac, and no affinity for the aldehyde, E6,Zll-16Ald, when the protein is 
incubated with equal amounts of the three sex pheromones. 5 

Earlier experiments based on electroantennogram (EAG) and single sensillum recordings (SSR) 
highlighted the extraordinary specificity and sensitivity of the insect olfactory system. It has been clearly 
demonstrated that each olfactory receptor neuron in a sensillum is highly tuned to a key stimulus (e.g., a 
pheromone constituent) such that minimal structural modification to a pheromone molecule renders it 
inactive. 13 The large number of sensilla distributed over the surface of the antennae and maxillary palps 
most likely contributes to the sensitivity of the insect olfactory system, but selectivity is mediated by 
molecular recognition at the periphery. Selectivity does not have to rely entirely on the odorant receptors, 
if odorant-binding proteins filter out some of the potential receptor ligands. 

Insects have evolved molecular mechanisms for the rapid inactivation or deactivation of chemical 
signals. To clear up the ORs and avoid continuous activation by “stray” semiochemicals, the signal 
from molecules that have already been conveyed must be terminated immediately. It has been 
suggested that the process is so rapid that it requires a hitherto unknown molecular mechanism for 
trapping the signal-carrying molecules (odorants). 14 On the other hand, it has also been demonstrated 
that inactivation can be accomplished by antennae-specific odorant-degrading enzymes. 15 ' 16 Indeed, 
inhibition of a pheromone-degrading enzyme in vivo led to the complete desensitization of highly 
sensitive, pheromone-specific olfactory receptor neurons in male antennae of a scarab beetle. 17 
It seems that localized low-pH environments generated by negatively charged surfaces on the 
dendrites are also essential to prevent “premature inactivation” of odorants. As demonstrated with 
ApolPDE, the pheromone-degrading enzyme of the wild silkmoth, Antheraea polyphemus, ODEs are 
fast at the bulk pH of sensillar lymph, but sluggish at the low pH environments where odorant 
“undocking” takes place. 16 The generalization of this finding must await further experiments given the 
diverse nature of insect ODEs. 

Because no odorant comes to the ORs except through OBPs, functional OBPs can be utilized as 
molecular targets for the screening of mosquito attractants and repellents in an approach similar to 
receptor-based drug discovery. While we have gained a better understanding of the molecular basis of 
attractant reception. The mode of action of deet and other mosquito repellents is not yet known. It is 
known, however, that deet has an olfactory-based repellent effect 18,19 as well as feeding-deterrent 
effect. 19 In addition, it has been demonstrated by single sensillum recordings, 20 gas chromatography 
coupled to antennographic detection 21 (Figure 11.3), and electroantennograms 22 that mosquito possess 
deet-detecting olfactory receptor neurons. These findings suggest that deet may be bound and transported 
by OBPs as has been shown for other odorants. Therefore, mosquito OBPs are likely suitable molecular 
targets for the development of repellents. 

Because ODE inhibitors desensitize the olfactory system, 17 an untapped strategy could exploit the 
design of anosmia-inducing ODE inhibitors to reduce mosquito bites. To inhibit ODEs in vivo, a 
compound must be assisted by odorant-binding protein to penetrate the sensillar lymph and reach the 
molecular target. Environmentally safe ODE inhibitors could be designed by a rational approach 


© 2006 by Taylor & Francis Group, LLC 


Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents 


233 




FIGURE 11 .3 Separation of a sample of deet by gas chromatography with simultaneous recording with a flame ionization 
detector (upper trace) and a mosquito antennae-based biosensor (lower trace). Antennae of blood-fed female Culex pipiens 
pallens were used for electroantennography (lower trace). The sample (10 pg) was injected in splitless mode and separated 
on a HP-5MS capillary column (30 m X 0.25 mm; 0.25 pm; Agilent Technologies, Palo Alto, CA) that was operated at 100°C 
for 1 min, increased to 270°C at a rate of 15°C/min, and held at this temperature for 10 min. Deet appeared at 8.43 min with 
an unambiguous EAD response. Smaller peaks at 7.4, 7.6, and 9.3 min are EAD-inactive chemical impurities. 


2006 by Taylor & Francis Group, LLC 



234 


Insect Repellents: Principles , Methods, and Uses 


considering structural features of the binding cavities of OBPs and catalytic sites of ODEs. By decreasing 
mosquito bites with novel and user-friendly repellents, disease transmission could also be reduced. 

Choosing Functional Molecular Targets 

The number of functional OBPs that exist for a single insect species is unknown, but to date only one 
pheromone-binding protein 23 ' 24 (and a few general odorant-binding proteins) have been identified from 
the silkworm moth, Bombyx more, whereas the malaria mosquito, Anopheles gambiae, for example, has 
potentially as many as 55 OBPs. 23 (Initially, 57 were suggested, but Biesmann and collaborators 26 found 
that OBP34 and OBP37 genes encode the same OBP and the proteins predicted from OBP35 and OBP36 
are identical.) This huge discrepancy in number of OBPs per species may be related to the method of 
“identification” of OBPs. Protein-based approaches are aimed at the isolation and identification of OBPs, 
followed by the cloning of the genes (or cDNAs) encoding these proteins. On the other hand, the gene- 
based approaches provide little data on expression and functions of proteins. While minor OBPs may be 
expressed at levels below the detection limits of the protein-based methods, the gene-based approach 
may lead to putative proteins which may not even be expressed in the sensillar lymph of insect antennae 
or maxillary palps. Even if a single OBP is involved in the detection of multiple compounds, one would 
expect that the insect antennae possess multiple OBPs since insects can detect a number of 
physiologically relevant compounds with diverse chemical structures, derived from conspecifics 
(pheromones), hosts, or potential ovipostion sites. However, it is highly unlikely that all OBPs predicted 
from an insect’s genome are indeed olfactory proteins. As an example, it has been suggested that the 
OBP-gene family of Drosophila melanogaster comprises as many as 51 putative OBPs, 27 but only seven 
of them have been demonstrated to be expressed specifically in olfactory organs of adults (antennae only 
or antennae and maxillary palps): Obpl9a, Obp57a, Obp69a (formerly named PBPRP-1), Obp83a 
(PBPRP-3, OS-F), Obp83b (OS-E), Obp84a (PBPRP-4), and Obp99d. Two other putative OBPs namely, 
Obp28a (PBPRP-5) and Obp76a (LUSH), are detected in the antennae of adults as well as in larval 
chemosensory organs. 28 The same is true for mosquito "OBPs” whose genes are more broadly 
expressed. 29 For example, out of 20 OBP genes, Li et al. 29 found three genes expressed in all tissues 
and three that are either expressed at low level or not expressed at all in adults. Biessmann and 
collaborators 26 employed microarray and real-time quantitative RT-PCR in an attempt to obtain a better 
understanding of the expression patterns of the genes possibly involved in reception of host odorants in 
females of Anopheles gambiae. Twenty-four “typical” OBP genes were detected above background 
levels, some with higher expression levels in female or male antennae; whereas, others were detected in 
antennae but not in maxillary palps, and others in both olfactory tissues. Conversely, RT-PCR analysis 
showed that genes suggested by microarray analysis to be expressed predominantly in antennae were also 
expressed in nonolfactory tissues, as well as in mosquito larvae. 26 

Olfactory and nonolfactory proteins from the OBP-gene family appear to belong to the same structural 
family. Their helix-rich structures suggest that these proteins encapsulate hydrophobic odorants and 
other ligands, with the ability to transport them in aqueous environments. 2 Therefore, proteins of this 
group should be named encapsulins to imply the common role of encapsulating small ligands. 2 The 
labeling “OBP” then should be restricted to olfactory odorant-binding proteins. It is possible that a large 
number of genes annotated from insect genomes as putative OBPs are merely encapsulins. One of the 
criteria widely utilized to annotate/identify putative OBPs is the occurrence of six well-conserved 
cysteine residues. The spacing patterns are also structurally important. While the six-cysteine pattern is a 
hallmark for most moth OBPs identified to date, it is not limited to OBPs. Insect defensins, for example, 
share the same feature. Conversely, not all odorant-binding proteins and antennae-specific proteins 
(putative OBPs) possess only six cysteine residues. For example, pheromone-binding proteins with 
seven-cysteine residues have been isolated from a silkmoth, Samia cynthia ricini (Leal, unpublished 
data) and predicted from the genome of Drosophila 27 and Anopheles gambiae. 30 We have also isolated 
and cloned an eight-cysteine, antennae-specific, putative OBP from the yellow fever mosquito. 


© 2006 by Taylor & Francis Group, LLC 



Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents 


235 


Aedes aegypti. 3 ' On the other hand, it is unlikely that putative OBPs with 12-cysteine residues deduced 
from the genomes of Drosophila 21 (Obp58b, Obp58c, Obp58d, Obp83c, Obp93a) and Anopheles 
gambiae 30 would bind, transport, and release ligands in the same way as described for pheromone¬ 
binding proteins. 

When using olfactory proteins as molecular targets in studies aimed at the design of potential mosquito 
attractants or repellents, it is essential to focus on functional OBPs. A solid literature on pheromones and 
sensory physiology has laid the foundation, in that it has revealed that moth pheromone-binding proteins 
are expressed specifically in male antennae and are restricted to pheromone-detecting sensilla. For 
example, long sensilla trichodea are present in both male and female antennae of the silkworm moth, 
Bombyx mori. In males, these sensilla respond to bombykol and bombykal 32 ' 33 ; whereas, in females they 
respond to benzoic acid and linalool. 34 By immunolocalization of different OBPs with specific antisera, 
Steinbrecht and collaborators showed that BmorPBP is expressed in pheromone-detecting sensilla 
trichodea. 35 In contrast, general odorant-binding proteins are detected in most sensilla basiconica, 36 a 
detector for plant-derived compounds. In addition, the female long sensilla trichodea, morphologically 
identical to the male pheromone detectors, express a general odorant-binding protein 37 and detects only 
nonpheromonal compounds. 

Mapping of mosquito sensilla on antennae and maxillary palps, and the identification of all 
physiologically relevant ligands (semiochemicals) are yet to be completed. Thus, mosquito OBPs can 
not be identified on the basis of the same gold standards as for moth OBPs. However, it is reasonable to 
assume that OBPs expressed in olfactory (antennae and/or maxillary palps), and not in nonolfactory 
tissues, are functional proteins (olfactory OBPs). In a recent structural study (see above), 11 we focused on 
an OBP from the malaria mosquito (AgamOBPl), which is expressed in antennae, but not in legs (control 
tissue) (Figure 11.4). Following the protocol that led to the isolation of the first OBPs from mosquito 
species 38,39 we isolated four antennae-specific proteins from Anopheles gambiae (Mopti strain) 
(Figure 11.4). MALDI-TOF (matrix-assisted laser desorption/ionization-time of flight) mass spectrometry 
and tandem (LC-MS-MS) mass spectral analysis of the protein bands (Agl—4) led to isolation of OBPs 
whose genes have been previously identified 40 and one putative odorant-degrading enzyme. Therefore, we 
focused on the main antennae-specific protein, AgamOBPl, in our structural biology studies (see above). 


Screening Techniques 

Prospecting for novel mosquito attractants or repellents can be based on molecular interactions of 
candidate compounds with olfactory proteins. Regardless of having odorant receptors or odorant-binding 
proteins as molecular targets, these screening techniques do not completely replace behavioral or field 
studies. However, large numbers of test compounds can be eliminated if they can not be transported to 
odorant receptors or do not activate these receptors. Provided that the appropriate molecular targets have 
been selected, these strategies can shorten the list of test compounds for further in-depth evaluations. 

Receptor-Based Approach 

In principle, the screening of potential mosquito attractants and repellents can be based on receptor- 
ligand interactions. This could be achieved either by heterologous expression of putative odorant 
receptors or in vitro binding assays. The latter is technically challenging, but the former could be 
performed, for example, with candidate receptors expressed in Xenopus oocytes. 12 Another promising 
avenue is the expression of target ORs in a mutant of Drosophila containing an “empty neuron.” The 
odorant receptors Or22a and Or22b of Drosophila were shown to be co-expressed specifically in the 
ab3A antennal neuron 41 and a mutant (A halo) lacking these genes has been utilized for functional 
analysis of odorant receptors of Drosophila. 42 Two odorant receptors from the malaria mosquito. 


© 2006 by Taylor & Francis Group, LLC 




236 


Insect Repellents: Principles , Methods, and Uses 



FIGURE 11.4 Gel electrophoresis (15% native) analysis of female antennae-specific proteins from Anopheles gambiae. 
Antennal (ANT; 700 antennae) and leg (100 hindlegs) extracts from 7-10 day-old Anopheles gambiae females (Mopti). The 
migration of a moth OBP (BmorPBP) is indicated by a bar above Ag2. Agl was identified as AgamOBPl. 


An. gambiae, have been expressed in the A halo mutant, with the response of the ab3A neuron of 
Drosophila being analyzed by single sensillum recordings. 43 Based on the response of AgOrl to 
4-methylphenol and AgOr2 to 2-methylphenol. these then putative odorant receptors were identified as 
the first odorant receptor genes from the malaria mosquito. 43 

It is possible that mutants of Drosophila engineered with malaria mosquito odorant receptors could be 
used to screen for other potential ligands of the AgORl and AgOR2 receptors. It is not certain, however, 
if the OBPs compartmentalized in the sensillum-housing the ab3A neuron are necessary and sufficient to 
mimic the olfactory system of the malaria mosquito. Further experiments may clarify if host-finding- 
related receptors in the mosquito will be functional in an engineered mutant of Drosophila. 

Binding Assay-Based Approach 

Binding of a semiochemical to an OBP can be investigated by incubation of a recombinant protein 
and test ligands, with binding affinity being assessed, for example, by fluorescence, calorimetry, or by 


© 2006 by Taylor & Francis Group, LLC 



Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents 


237 


measuring the amount of bound ligand. Intrinsic protein fluorescence can be very sensitive and 
requires low amounts of protein and ligand, 4 but signal-to-noise may be too small if there is no ligand- 
induced conformational change leading to a change in the environment of tryptophan residues. This 
difficulty can be overcome by employing a reporter group either as a ligand 44 or by attaching a 
fluorophore to the protein in a sensitive environment. 45 The use of extrinsic reporter groups may 
enhance sensitivity dramatically, but it may be difficult to determine if the labeling is not affecting the 
normal function of the protein. This is particularly problematic for OBPs whose structures and cognate 
ligands are not yet known, which is the case for mosquito OBPs. With a noncovalent fluorescent probe 
the protein may be tested in its native conformation. These probes are normally nonfluorescent in 
water, but their emission spectra are modified when bound to a protein normally undergoing a blue 
shift with a marked increase in intensity. Binding of a test ligand to an OBP can be determined by 
quenching the extrinsic fluorescence. The most widely used probes are 1-aminoanthracene (1-AMA) 
and A-phenyl-l-naphthylamine (1-NPN), which were initially employed in binding experiments with 
vertebrate OBPs. 46 

Isothermal titration calorimetry (ITC) has been employed to demonstrate binding of 2-isobutyl-3- 
methoxypyrazine to an OBP (ASP2) from the honeybee. 47 Despite several attempts, we were unable to 
measure binding of bombykol to BmorPBP by ITC (Leal, unpublished data). Preliminary attempts to 
employ surface plasmon resonance (BIACORE®*) were also unrewarding, probably because of both the 
low solubility and small size of the ligand (analyte in BIACORE jargon) (Leal, unpublished data). 

In early work to identify pheromone-binding proteins, radiolabeled pheromones were employed in 
qualitative binding assays. With the availability of recombinant proteins, radiolabeled pheromones can 
be used in quantitative assays in which free ligands are separated by gel filtration from bound ligands. 48 
Although a valuable tool for studies of pheromone-PBP interactions, this type of binding assay has 
limited application in screening programs because radioactive test ligands are required. If a library of 
radioactive test compounds were available, one might as well employ a high-throughput screening, such 
as the scintillation proximity assay. 49 

Recently, we have developed a low-throughput screening protocol named the cold binding assay 10 
(Figure 11.5) that does not require radioactive (hot) ligands. After incubation of test compound(s) with an 
OBP, the free ligand is removed by filtration, whereas the protein bound ligand is retained in the 
centrifugal device and extracted with an internal standard containing organic solvent. Binding is 
quantified by gas chromatography and the identity of the recovered ligand is confirmed by gas 
chromatography-mass spectrometry. As a negative control, binding is also investigated at low pH. 
Among other advantages, this protocol allows competitive binding assays 5 in which the best ligand can 
be determined in a single assay. 

A promising strategy for online screens is the covalent immobilization of OBP to a liquid 
chromatography stationary phase. 50 We have tested this principle with two odorant-binding proteins, 
the PBP from the silkworm moth, BmorPBP, 23 ' 24 and an OBP from Culex quinquefasciatus, 
CquiOBPl. 38 The BmorPBP column distinguished four compounds, with bombykol showing the 
highest affinity, followed by bombykal, 1-hexadecanol, and (Z,£j-5,7-dodecadien-l-ol. 50 Zonal 
chromatographic studies using D-, L-, and D/L-lactic acid showed that the CquiOBPl column separated 
the two isomers of lactic acid, with L-lactic acid having higher affinity. 

Validating Molecular Targets 

The reverse chemical ecology approach described above has already been employed for the development 
of better lures for the Navel Orangeworm moth, Amyelois transitella. 51 ' 52 Despite the tremendous effort 
by leading scientists in the field of chemical ecology, only one constituent of the pheromone system of 
this species was known until recently. 53 With a multidisciplinary approach, including OBP-based 


A registered trademark of Biacore International AB, Uppsala, Sweden. 


© 2006 by Taylor & Francis Group, LLC 



238 


Insect Repellents: Principles, Methods, and Uses 


Cold Binding Assay 


1. Incubation -- ,- 2. Separation 



FIGURE 11.5 Schematic view of the four steps of a cold binding assay. A glass insert A deactivated by Silcote CL7 
treatment (Kimble Chromatography, Vineland, NJ) is used to incubate protein and test ligand B. The reaction mixture is 
shaken (100 rpm) at 25 + 2°C for 1 h C. For separation of the bound and free ligands, the reaction mixture is transferred to a 
washed Microcon YM-10 (Millipore) D and centrifuged (12,000Xg, 4°C) for 5 min E. The retentate is transferred to a 100 pi 
V-vial (Wheaton, Millville, NJ) F along with a hexane containing an internal standard (eicosyl acetate, Fuji Flavor Co., 
Tokyo, Japan). The vial is capped G, vortexed for 1 min, and then centrifuged (2,500 X g, 4°C) for 5 min. The hexane fraction 
(upper layer) is recovered H and analyzed by gas chromatography (GC) for quantification I. The extract can be analyzed by 
gas chromatography-mass spectrometry (GC-MS) to confirm identification of the ligand extracted from the OBP-ligand 
complex. 


screening of potential attractants, we discovered a complex pheromone system. 51 ' 52 In addition, a 
mosquito OBP-based screening program aimed at the development of oviposition attractants is underway 
in my laboratory. 

The success of these programs depends heavily on the utilization of appropriate molecular targets, i.e., 
functional olfactory proteins. Our strategy is to select major antennae-specific olfactory proteins based on 
the assumption that the most abundant OBPs in mosquito antennae play critical roles in chemical 
communication as PBPs do in moths. Two lines of evidence substantiate this hypothesis. Based on the 
binding of the mosquito oviposition pheromone (MOP) 54 to an OBP, previously isolated from Culex 
quinquefasciatus , 38 we now have evidence that a major female antennae-specific protein in Culex 
quinquefasciatus plays a critical role in insect olfaction (Leal et al., unpublished data). In addition, 
sensilla in female antennae that are tuned to MOP house other olfactory receptor neurons sensitive to 
chemical cues used for attraction to oviposition sites (Syed and Leal, unpublished data). A second line of 
evidence comes from binding studies with an OBP previously isolated from the antennae of the yellow 
fever mosquito, Aedes aegypti, AaegOBPl. 31 Binding assays (Figure 11.6) indicate that at pH 7 
AaegOBPl bound nonanal, whereas no binding was found at low pH. Nonanal is the active ingredient 


© 2006 by Taylor & Francis Group, LLC 













































Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents 


239 


15 - 



FIGURE 11.6 Nonanal binds with high affinity to AaegOBPl at pH 7. The amount of ligand recovered at low pH 5 is not 
significantly different from the amount detected in buffer (JV=5), indicating no affinity at low pH. 

of a commercially available lure (AtrAedes®*) utilized for monitoring populations of gravid females of 
Aedes aegypti. 


Concluding Remarks 

The state-of-the-art screening programs described here should not be oversold as the panacea for 
controlling mosquito-borne diseases. To generate practical applications, these molecular-based 
screening strategies have to be integrated with sensory physiology, behavioral bioassays, and held 
studies. Interaction of a ligand with an olfactory protein does not necessarily imply full physiological 
function, behavioral response, mosquito trapping, or reduced biting. While we hope that these molecular- 
based programs may ultimately lead to the decrease of mosquito-transmitted diseases, the discovery of 
new repellents and attractants is just a stepping stone towards the ultimate goal. 

Acknowledgments 

This work was supported by the NIH-National Institute of Allergy and Infectious Diseases 
(1U01AI058267-01), a Specific Cooperative Agreement (No. 58-1275-1-042) with Chemical Affecting 
Insect Behavior Laboratory, Agricultural Research Service, U.S. Department of Agriculture, and a 
Research Agreement with Bedoukian Research, Inc. I benefited greatly from discussions with past and 
current undergraduate, graduate, postdoctoral students, and visiting scientists in my laboratory as well as 
with various collaborators over the years. I thank Yuko Ishida, Zainulabeuddin Syed, and Wei Xu for 
their suggestions to improve an earlier draft of the manuscript; Wei Xu for running the binding assay with 
AaegOBPl, and Dr. Greg Lanzaro for providing Anopheles gambiae mosquitoes for protein extraction. 
Helpful comments and suggestions by Dr. George Kamita, Dr. Mark Wogulis and my department 
colleague and collaborator. Dr. Anthon Cornel, are also highly appreciated. 


Registered trademark of Ecovec, Ltd, Brazil. 


© 2006 by Taylor & Francis Group, LLC 













240 


Insect Repellents: Principles, Methods, and Uses 


References 

1. J. Murlis, M. A. Willis, and R. T. Carde, Spatial and temporal structures of pheromone plumes in fields 
and forests, Physiol. Entomol., 25, 211-222, 2000. 

2. W. S. Leal, Proteins that make sense, in Insect Pheromone Biochemistry and Molecular Biology, The 
Biosynthesis and Detection of Pheromones and Plant Volatiles, G. J. Blomquist and R. G. Vogt (Eds.), 
London: Elsevier Academic Press, 2003, pp. 447-476. 

3. W. S. Leal, Pheromone reception. Top. Curr. Chem., 240, 1-36, 2005. 

4. H. Wojtasek and W. S. Leal, Conformational change in the pheromone-binding protein from Bombyx 
mori induced by pH and by interaction with membranes, J. Biol. Chem., 274, 30950-30956, 1999. 

5. W. S. Leal, A. M. Chen, and M. L. Erickson, Selective and pH-dependent binding of a moth 
pheromone to a pheromone-binding protein. J. Chem. Ecol., 31, 2493-2499, 2005. 

6. B. H. Sandler et al.. Sexual attraction in the silkworm moth: Structure of the pheromone-binding- 
protein-bombykol complex, Chem. Biol, 7, 143-151, 2000. 

7. F. Damberger et al., NMR characterization of a pH-dependent equilibrium between two folded 
solution conformations of the pheromone-binding protein from Bombyx mori. Protein Sci., 9, 
1038-1041, 2000. 

8. W. S. Leal, Duality monomer-dimer of the pheromone-binding protein from Bombyx mori, Biochem. 
Biophys. Res. Commun., 268, 521-529, 2000. 

9. R. Horst et al., NMR structure reveals intramolecular regulation mechanism for pheromone binding 
and release, Proc. Natl. Acad. Sci., USA, 98, 14374-14379, 2001. 

10. W. S. Leal et al., Kinetics and molecular properties of pheromone binding and release, Proc. Natl. 
Acad. Sci., USA, 102, 5386-5391, 2005. 

11. M. Wogulis et al.. The crystal structure of an odorant binding protein from Anopheles gambiae: 
Evidence for a common ligand release mechanism, Biochem. Biophys. Res. Commun., 339, 157-164, 
2006. 

12. C. H. Wetzel et al., Functional expression and characterization of a Drosophila odorant receptor in a 
heterologous cell system, Proc. Nat. Acad. Sci., USA, 98, 9377-9380, 2001. 

13. K. -E. Kaissling, R.H. Wright Lectures on Insect Olfaction, British Columbia: Simon Fraser University, 
1987. 

14. K. -E. Kaissling, Olfactory perireceptor and receptor events in moths: A kinetic model, Chem. Senses, 
26, 125-150, 2001. 

15. R. G. Vogt, L. M. Riddiford, and G. D. Prestwich, Kinetic properties of a sex pheromone-degrading 
enzyme: The sensillar esterase of Antheraeapolyphemus, Proc: Natl. Acad. Sci., USA, 82, 8827-8831, 
1985. 

16. Y. Ishida and W. S. Leal, Rapid inactivation of a moth pheromone, Proc: Natl. Acad. Sci., USA, 102, 
14075-14079, 2005. 

17. M. Maibeche-Coisne et al., Pheromone anosmia in a scarab beetle induced by in vivo inhibition of a 
pheromone-degrading enzyme, Proc. Natl. Acad. Sci., USA, 101, 11459-11464, 2004. 

18. E. J. Hoffmann and J. R. Miller, Reassessment of the role and utility of wind in suppression of mosquito 
(Diptera: Culicidae) host finding: Stimulus dilution supported over flight limitation, J. Med. Entomol, 
40, 607-614, 2003. 

19. J. A. Klun, A. Khrimian, and M. Debboun, Repellent and deterrent effects of SS220, Picaridin, and 
DEET suppress human blood feeding by Aedes aegypti. Anopheles stephensi, and Phlebotomus 
papatasi, J. Med. Entomol, 43, 34-39, 2006. 

20. J. Boeckh et al., Acylated 1,3-aminopropanols as repellents against bloodsucking arthropods. Pest. Sci., 
48, 359-373, 1996. 

21. W. S. Leal and K. Uchida, Application of GC-EAD to the determination of mosquito repellents derived 
from a plant, Cymbopogon citratus, J. Asia-Pacific Entomol., 1, 217-221, 1998. 

22. C. Constantini et al., Electroantennogram and behavioural responses of the malaria vector Anopheles 
gambiae to human-specific sweat components, Med. Vet. Entomol., 15, 259-266, 2001. 

23. R. Maida et al.. The pheromone binding protein of Bombyx mori: Purification, characterization and 
immunocytochemical localization. Insect Biochem. Molec: Biol., 23, 243-253, 1993. 


© 2006 by Taylor & Francis Group, LLC 



Molecular-Based Chemical Prospecting of Mosquito Attractants and Repellents 


241 


24. J. Krieger et al., Binding proteins from the antennae of Bombyxmori, Insect Biochem. Molec. Biol., 26, 
297-307, 1996. 

25. P. X. Xu, L. J. Zwiebel, and D. P. Smith. Identification of a distinct family of genes encoding atypical 
odorant-binding proteins in the malaria vector mosquito. Anopheles gambiae. Insect Mol. Biol., 12, 
549-560, 2003. 

26. H. Biessmann et al.. Microarray-based survey of a subset of putative olfactory genes in the mosquito 
Anopheles gambiae. Insect Mol. Biol., 14. 575-589, 2005. 

27. D. S. Hekmat-Scafe et al., Genome-wide analysis of the odorant-binding protein gene family in 
Drosophila melanogaster. Genome Res., 12, 1357-1369, 2002. 

28. K. Galindo and D. P. Smith. A large family of divergent Drosophila odorant-binding proteins expressed 
in gustatory and olfactory sensilla. Genetics, 159, 1059-1072, 2001. 

29. Z. X. Li et al., Identification and expression of odorant-binding proteins of the malaria-carrying 
mosquitoes Anopheles gambiae and Anopheles arabiensis. Arch. Insect. Biochem. Physiol., 58, 
175-189, 2005. 

30. J. J. Zhou et al., "Plus-C” odorant-binding protein genes in two Drosophila species and the malaria 
mosquito Anopheles gambiae. Gene, 327, 117-129, 2004. 

31. Y. Ishida et al.. Intriguing olfactory proteins from the yellow fever mosquito, Aedes aegypti, 
Naturwissenschaften, 91, 426-431, 2004. 

32. K.-E. Kaissling et al., A new pheromone of the silkworm moth Bombyx mori. Sensory pathway and 
behavioral effect, Naturwissenschaften, 65, 382-384, 1978. 

33. K. -E. Kaissling and E. Priesner, Die Riechschwelle des Seidenspinners, Natunvissenschaften, 57, 
23-28, 1970. 

34. E. Priesner, Progress in the analysis of pheromone receptor systems, Ann. Zool. Ecol. Anim., 11, 
533-546, 1979. 

35. R. A. Steinbrecht. M. Ozaki, and G. Ziegelberger. Immunocytochemical localization of pheromone¬ 
binding protein in moth antennae, Cell Tissue Res., 270, 287-302, 1992. 

36. M. Laue, R. A. Steinbrecht, and G. Ziegelberger, Immunocytochemical localization of general odorant¬ 
binding protein in olfactory sensilla of the silkmoth Antheraea polyphemus, Naturwissenschaften, 81, 
178-180, 1994. 

37. M. Laue and R. A. Steinbrecht, Topochemistry of moth olfactory sensilla, Int. J. Insect Morphol. 
Embiyol., 26, 217-228, 1997. 

38. Y. Ishida, A. J. Cornel, and W. S. Leal. Identification and cloning of a female antenna-specific 
odorant-binding protein in the mosquito Culex quinquefasciatus, J. Chem. Ecol., 28, 867-871, 
2002 . 

39. Y. Ishida, A. J. Cornel, and W. S. Leal. Odorant-binding protein from Culex tarsalis, the most 
competent vector of West Nile Virus in California, J. Asia-Pacific Entomol., 6, 45-48, 2003. 

40. H. Biessmann et al., Isolation of cDNA clones encoding putative odourant binding proteins from 
the antennae of the malaria-transmitting mosquito. Anopheles gambiae. Insect Mol. Biol., 11, 
123-132, 2002. 

41. A. A. Dobritsa et al.. Integrating the molecular and cellular basis of odor coding in the 
Drosophila antenna. Neuron, 37, 827-841, 2003. 

42. E. A. Hallem, M. G. Ho. and J. R. Carlson, The molecular basis of odor coding in the 
Drosophila antenna. Cell, 117. 965-979. 2004. 

43. E. A. Hallem et al., Mosquito receptor for human-sweat odorant. Nature, 427, 212-213, 2004. 

44. L. Ban et al., Binding properties of a locust’s chemosensory protein, Biochem. Biophys. Res. 
Commun., 293, 50-54, 2002. 

45. N. Honson et al.. Structure—activity studies with pheromone-binding proteins of the gypsy moth, 
Lymantria dispar, Chem. Senses, 28, 479-489, 2003. 

46. P. Pelosi, et al.. Soluble proteins in insect chemical communication. Cell. Mol. Life Sci., in press. 

47. L. Briand et al.. Ligand binding and physico-chemical properties of ASP2, a recombinant 
odorant-binding protein from honeybee ( Apis mellifera L.), Eur. J. Biochem., 268, 752-760, 2001. 

48. E. Plettner et al.. Discrimination of pheromone enantiomers by two pheromone binding proteins 
from the Gypsy moth Lymantria dispar, Biochemistry, 39, 8953-8962, 2000. 


2006 by Taylor & Francis Group, LLC 



242 


Insect Repellents: Principles, Methods, and Uses 


49. N. D. Cook, Scintillation proximity assay: A versatile high-throughput screening technology, 
Drug Discov. Today, 1, 287-294, 1996. 

50. A. Margaryan, et al., Synthesis of an immobilized Bombyx mori pheromone binding protein liquid 
chromatography stationary phase, Talanta, in press. 

51. W. S. Leal et al., Unsual pheromone chemistry in the navel orangeworm: Novel sex attractants 
and a behavioral antagonist, Naturwissenschaften, 92, 139-146, 2005. 

52. W. S. Leal et al.. Molecular-based screening of attractants for the Navel Orangeworm, in Proc. 
33rd Almond Industry Conference. Modesto, CA, 27-35, 2005. 

53. J. A. Coffelt et al., Isolation, identification, and synthesis of a female sex pheromone of the navel 
orangeworm, Amyelois transitella (Lepidoptera: Pyralidae), J. Chem. Ecol., 5, 955-966, 1979. 

54. B. R. Laurence and J. A. Pickett, £Vyf/tro-6-Acetoxy-5-hexadecanolide, the major component of a 
mosquito oviposition attractant pheromone, J. Chem. Soc. Chem. Commun., 59-60, 1982. 

55. R. A. Roque and A. E. Eiras, Avaliagao de atraente sintetico de oviposigdo para femeas gravidas 
de Aedes aegypti (Diptera: Culicidae), in IV Encontro Brasileiro de Ecologia Quunica, 
Piracicaba: SP-Brazil, 156, 2005. 

56. C. Lautenschlager, W. S. Leal, and J. Clardy, Coil-to-helix transition and ligand relase of Bombyx 
mori pheromone-binding protein, Biochem. Biophys. Res. Commun., 335, 1044-1050, 2005. 


© 2006 by Taylor & Francis Group, LLC 



12 


Evaluation of Topical Insect Repellents and Factors 
That Affect Their Performance 


Scott P. Carroll 


CONTENTS 

Introduction.245 

History.246 

Types of Tests—Background.247 

Factors Affecting Repellent Performance.248 

Mosquito Taxonomy and Genetics.248 

Individual Human-Subject Differences.249 

Conditions of Use.250 

Formulation Chemistry.251 

Active Ingredients and Their Efficacy Assessment.252 

Laboratory Efficacy Comparisons.252 

Field Efficacy Comparisons.253 

Conclusions.255 

Acknowledgments.256 

References.256 


Introduction 

Personally-applied topical insect repellents are a flexible and relatively affordable means of gaining 
protection from biting arthropods and the disease-causing pathogens they sometimes carry. 1 ' 2 Although a 
number of useful repellents have been developed, a variety of factors limits their effectiveness in 
application. The purpose of this chapter is to review those factors, consider their importance, and discuss 
means of overcoming them. The majority of investigations have been conducted against mosquitoes that 
are vectors of important disease agents: the yellow fever mosquito, Aedes aegypti, and the Anopheles 
species that transmit malaria pathogens. Although this chapter emphasizes results from studies of 
mosquitoes, data from other biting arthropods are included when helpful or relevant. 

For a repellent to be successful, it must first have a high percentage of effectiveness against the biting 
arthropods of concern for the entire period of likely use. Second, it should be toxicologically safe at the 
rate of application for which it is intended. Third, it should be easy to apply and pleasant on the skin in 


245 


© 2006 by Taylor & Francis Group, LLC 





















246 


Insect Repellents: Principles , Methods, and Uses 


terms of residual feeling and odor. Finally, the entire spectrum of costs involved in production and 
marketing of the repellent should result in a product that is reasonably priced for the consumer. Among 
the repellent active ingredients formulated over the last half century, deet (AUV-diethyl-3-methylbenza- 
mide) has been included in numerous products that come remarkably close to approaching that ideal, and 
it is estimated that deet is employed at least 200 million times per year around the globe. 3 Persistent 
public concerns about its safety (some based on hearsay) have been aggravated by its cosmetic 
shortcomings and plasticizing (i.e., tendency to soften plastics) effects. Cosmetic improvements have 
been achieved mainly by limiting deet concentrations to 10% or lower, resulting in formulations with 
efficacy of limited duration. In addition, while high-concentration deet formulations often remain 
efficacious for eight or more hours, attempts to enhance duration by manipulating carrier formulations 
have not resulted in substantial improvements. This suite of concerns has helped to fuel the search for 
suitable alternatives for both civilian and military applications. 

Little is known about how insect repellents function. 4 Such knowledge would promote the 
development of more effective repellents based on biochemical and neurophysiological principles. In 
the absence of real knowledge about mechanisms, we may instead progress inferentially through the 
collation and analysis of natural history data on factors that influence success. Interactions between 
parasites and hosts are biologically complex and therefore inherently dynamic and challenging to 
control. Among the many factors likely to influence the effectiveness of a repellent are those involving 
the active ingredient and formulation, biology of the arthropod, the conditions and mode of use, and 
lastly, individual user traits. The diversity of variables and their interactions makes the precise 
measurement of performance difficult, requiring a great deal of empirical effort. Organized testing 
schemes that control variables systematically are therefore especially useful. Nonetheless, the 
complexity of host-parasite interplay suggests a priori that protection afforded by even the best active 
ingredient in an ideal formulation is likely to differ among arthropod taxa and among individual human 
subjects. Accordingly, comparative studies that examine such interactions should be especially valuable 
for advancing repellent science. 

In spite of these challenges, a number of promising active ingredients and formulation technologies 
have recently been developed. By identifying the liabilities that influence repellent performance, chances 
are now better than in the past to integrate the new resources to create superior, longer-lasting, more 
universally acceptable insect repellents. Laboratory tests are effective for screening purposes and for 
making comparisons under controlled conditions. Field tests give a better picture of repellent 
performance in actual use, and highlight the importance of the environment and other conditions of 
use. Accordingly, this chapter first reviews studies that describe the action and importance of factors that 
influence repellent performance. It then considers those factors in evaluating recent performance tests of 
promising deet alternatives. The goal is to present information that is directly relevant to issues faced by 
contemporary decision makers and to emphasize the importance of recognized variables, the better 
understanding of which may improve development prospects. 


History 

Insect repellents have been examined systematically in the U.S. since World War II, when military 
initiatives, in response to outbreaks of malaria in American soldiers in tropical theaters, were taken up by 
the U.S. Department of Agriculture (USDA). 5 That work mainly involved the screening of novel active 
ingredients against caged laboratory populations of Aedes aegypti and Anopheles albimanus. 6 
Ultimately, however, substantial work also assessed factors that influenced the performance of known 
repellents (principally dimethyl phthalate and deet), particularly with regard to the duration of 
repellency. 7 Those pioneering studies established the fundamental importance of dosage and rate of 
loss for determining the period of protection. Among the chief factors they identified as influencing loss 


© 2006 by Taylor & Francis Group, LLC 



Evaluation of Topical Insect Repellents and Factors That Affect Their Performance 


247 


were rates of evaporation and absorption that differed among individuals, and abrasion by clothing. 
Individual attractiveness to a biting arthropod was also important, but gender, hairiness, sweat, and 
chemical deterioration were thought not to influence repellency. 7 While conceptually robust and 
comprehensive, most early studies had five or fewer subjects and probably served later researchers 
more in terms of intellectual guidance than through the specific applications of the results. 

In the succeeding four decades, basic research on repellents in the U.S. has continued to be sponsored 
heavily by the military and the USDA, with emphasis on extending duration. Industrial research over this 
period has stressed user acceptability and marketing appeal, whereas in Europe the market has more 
frequently addressed safety. Developing countries seem to stress cost (including searches for natural 
products). The majority of military work has been conducted with deet and laboratory strains of Aedes 
aegypti, although more recent work includes significant field studies and tests of experimental active 
ingredients. That initiative includes several studies that compared the original U.S. Army Insect 
Repellent (75% deet in ethanol, hereinafter “Army 75% deet”) to two polymerized deet lotions, 
specifically the 3M 34% deet formulation currently known as EDTIAR (extended duration topical 
insect and arthropod repellent) and marketed to the public as 3M Ultrathon"’, and the Biotek® 42% deet 
formulation. Such work is discussed in detail later in this chapter when the influence of formulation 
is considered. 


Types of Tests—Background 

Performance evaluations of repellents fall into two basic classes or design types. In the first approach, 
developed for field testing, a treated surface is exposed until a conservative, predefined failure event 
occurs, e.g., the time of the first bite, or the “first confirmed bite” (defined as the first bite that is followed 
by another bite within 30 min). This approach has the practical advantage of minimizing subject risk 
from wild mosquito bites. However, its scientific disadvantages include that the data set is truncated and 
minimized in size, and offers no basis for analyzing or comparing the period of partial protection after the 
onset of biting. In addition, truncation may inherently oversample that portion of the mosquito 
population that is most insensitive to the repellent. As pointed out by Rutledge in a number of 
publications, 8 measurements made of extreme individuals will be less reliable than those taken closer 
to the center of the population distribution. Depending on biting rates, some of these problems may be 
partially ameliorated by instead defining effective repellency as the duration of some percentage of 
protection (e.g., 90 or 95%) relative to the control. 

In field studies, an important factor influencing protection time is therefore likely to be the population 
size of the arthropod. 8 ' 9 Khan et al. 10 and Barnard et al. 11 reached similar conclusions based on 
experimental manipulations of mosquito numbers in cages. The probability that a test subject will 
encounter extremely insensitive arthropods will be higher in large parasite populations. Based on these 
statistical observations, Rutledge et al. 8 recommended the adoption of dose-response test design focused 
at the more typical portion of the mosquito population. At the median dose (i.e., the quantity required to 
repel 50% of the test arthropods), the result is essentially independent of the population size. Known as 
ED 50 (the minimum effective dosage to repel half of the arthropods), this test design allows much greater 
precision in the generation of a true estimate of repellent performance because of the inherent 
mathematics of error around a log-dose/probit curve. It also permits measurement of the sensitivity of 
different percentiles to population size, and focuses on percentiles of specific interest. 

"Minimum effective dosage” design and analysis is employed in laboratory evaluations of inherent 
repellency where the size of the test population is known. The resulting precision may be especially 


Ultrathon is a registered trademark of the 3M Corporation, Minneapolis, MN; Biotek is a registered trademark of Biotek 
Corporation, Woburn, MA. 


© 2006 by Taylor & Francis Group, LLC 





248 


Insect Repellents: Principles, Methods, and Uses 


valuable for comparing active ingredients and formulations. To bolster data quality and information 
content, field evaluations would likewise benefit from more extended records of biting events (i.e., 
extending the trial past the time of the first confirmed bite). For field testing, an important corollary of 
the foregoing is that the number of study subjects will directly influence the number of mosquitoes 
sampled, and thus the effective population size of mosquitoes from which data are collected. The 
common practice of employing just a few subjects per formulation (below) may therefore give a poor 
indication of the range of experiences that would characterize a larger sample of subjects. In other 
words, while analytical precision is gained from ED 50 laboratory studies by reducing the influence of 
rare insensitive mosquitoes, field evaluations of effective repellency benefit from the inclusion of 
exceptional mosquitoes, the avidity of which exceeds the capacity of the repellent to stop them from 
biting. It is important to sample with sufficient intensity to gauge performance against a large number 
of potentially biting individuals. 


Factors Affecting Repellent Performance 
Mosquito Taxonomy and Genetics 

The first comprehensive study of the interaction between repellency and mosquito taxonomy was 
conducted by Travis, 12 who showed that the ranking of protection provided by four repellents was not the 
same among two Aedes and two Anopheles species. Rutledge and colleagues conducted both intensive 8 
and extensive 13 studies of such interaction. In a study examining deet alone against Anopheles, Aedes, 
and Culex, the range in ED 50 was seven-fold. 8 Three species of Anopheles ranged from nearly the most, 
to the least easily repelled as a function of dosage. Even within a species (among ten strains of Aedes 
aegypti), they found significant variation in efficacy. Later, in a comparison of 31 repellent compounds, 
there was little or no predictability in performance rank across species. 13 Variation in observed 
repellency between species within a genus was as great as variation between species in different 
genera. Performance against Aedes aegypti was a poor predictor of performance against other 
mosquitoes, especially Anopheles species. 

In a series of incisive analyses, Curtis et al. 14 considered the interactions of mosquito species, 
repellents, and individual subject effects. Six species of mosquitoes from Anopheles, Aedes, and Culex 
were exposed to six repellents. The ED 50 of the repellents varied within and among genera by a factor 
ranging from three to 20-fold. Subjects differed in attractiveness, but not consistently across species of 
mosquitoes (assessed in the next section). Performance depended on the interaction of subject, repellent, 
and mosquito taxon. Similarly, Badolo et al. 15 found a repellent-by-taxon interaction in effective dosage 
of deet and Picaridin against native West African strains of caged Aedes aegypti and Anophles gambiae. 
Results from studies such as these discourage the notion that accurate performance generalizations are 
possible from tests with small numbers of subjects against a limited set of target species. 

Finally, Coleman et al. 16 broadened systematic comparisons further when considering the influence of 
deet, a lactone, and two piperidines against four Anopheles species, and two phlebotomines, Phlebotomus 
papatasi and Lutzomyia longipalpus. In general, Anopheles stephensi and the phlebotomines were the 
most susceptible to the repellents, and Anopheles albimanus was the least susceptible. Beyond those 
patterns, however, the relationship of performance among all the taxa was highly variable. Note also that 
deet is not always a superior repellent for phlebotomines. 17 

Given the high intergeneric, interspecific, and intraspecific variation in response to repellents observed 
in controlled laboratory settings, it is not surprising that the response has a genetic basis. Rutledge et al. 18 
established that repellent tolerance in Aedes aegypti is heritable, and in the case of deet involves partial 
dominance (one or a few genes of major effect). Such genetic control could result in an initially rapid 
phenotypic response to selection for deet tolerance. 


© 2006 by Taylor & Francis Group, LLC 



Evaluation of Topical Insect Repellents and Factors That Affect Their Performance 


249 


Individual Human-Subject Differences 

Bernier et al. (Chapter 4) reviewed the influence of human skin emanations on mosquito host location. 
Gilbert et al. 19 examined the influence of ten “subject variables” on attractiveness and repellency: 
gender, age, weight, skin temperature, skin moisture production, menses (females), and race, plus hair 
color and complexion within Caucasians. A remarkable sample size—50 adults of each gender—gave 
unusual statistical power to analyze subtle effects. The attractiveness tests were conducted in 
“olfactometer cages,” in which Aedes aegypti were exposed to air pulled across the surface of the 
repellent-treated arms of the subjects. The mosquitoes had the option of moving toward the arm and 
becoming trapped (and counted) as they approached it. Repellency was scored using 5% deet with 
exposure to mosquitoes at intervals. 

Only the effect of gender was clearly and strongly significant. On a proportional scale, the 
attractiveness of women was just 73% that of men. Only about 5% were more attractive than the 
male median. So while a few women were highly attractive to the mosquitoes (two of the ten most 
attractive subjects), all ten of the ten least attractive subjects were female. In terms of repellency, the 
lower female attractiveness was reflected in a 37% greater mean protection time for females as a group. 
Nonetheless, there was no significant correlation between individual attractiveness and protection time in 
either gender, suggesting that other factors are involved in repellent performance. 

Among the other factors investigated, subjects with the highest skin temperatures were more 
attractive or more poorly protected than those at the opposite extreme. Women with the highest 
moisture production from the skin were also more attractive than the opposite extreme, but that 
comparison yielded the reverse in men. Neither of these variables correlated with attractiveness or 
repellency across all subjects in a gender, however. Age, weight, menses, hair color and complexion 
were all inconsequential, 19 and the number of non-Caucasians tested was insufficient for meaningful 
interpretation of racial effects. No formal multivariate analyses of the dependent variables 
were conducted. 

Given the clarity of that study’s conclusion that women were less attractive and better protected 
from Aedes aegypti by deet, it is intriguing that a recent major study with Anopheles stephensi 
reported the opposite result. Golenda et al. 20 examined the duration of protection by EDTIAR to 
caged Anopheles stephensi in 60 female and 60 male volunteers. Self-dosing was performed by 
subjects in accordance with product label directions, and the mean rate of application was slightly 
higher in females (6%), but not significantly different from males. Biting rates on untreated arms were 
also the same between the sexes. Protection rates (relative to the untreated arms) are shown for each 
3-h sample interval in Table 12.1. Women experienced significantly less protection over time than 
did men. 

Examining an additional aspect of subject variation, Curtis et al. 14 reported that each subject’s relative 
attractiveness to mosquitoes is species-specific. Using caged Anopheles coustani , Culex quinquefas- 
ciatus, and Mansonia species, they found no predictable relationship between how the biting rate 


TABLE 12.1 

Comparative Repellency ((1—Biting Rate Treated)/!Biting Rate Control) X 
100) of U.S. Military EDTIAR (34% deet) on Male and Female Subjects 


Mean Repellency (%) 


Gender 

Oh 

3 h 

6 h 

9 h 

12 h 

Females 

100 

99.3 

92.8 

79.7 

66.3 

Males 

100 

100 

97.6 

91.9 

77.5 


Source: From C. F. Golenda, V. B. Solberg, R. B. Burge, J. M. Gambel, and R. A. 
Wirtz, American Journal of Tropical Medicine and Hygiene , 60, 654-657, 1999. 


© 2006 by Taylor & Francis Group, LLC 






250 


Insect Repellents: Principles , Methods, and Uses 


individuals experienced ranked from one species versus another. In addition to the possible effects of skin 
temperature and moisture, 19 ' 21 or their correlates, such inter-individual variation in attractancy may be 
influenced by differences in skin surface lipids. 22 Subjects may also vary in repellent performance due to 
differences in dermal absorption of the active ingredient, which in one study ranged from four to 14% of 
deet applied in a 15% ethanol solution. 23 

Conditions of Use 

Insect repellents are used in nature, where conditions may interact with user activity to influence 
repellency. It is well known that mosquitoes are most active under particular environmental conditions, 
and while optima vary among species, warm humid conditions with moderate to low light levels and low 
wind generally enhance mosquito foraging activity. Within the range of conditions appropriate for 
mosquito foraging, variation in temperature and humidity may not strongly influence biting rate and 
repellent performance. 24 Comparatively less is known about the state-dependence of mosquito foraging 
decisions beyond basic effects of age and parity. 25 For example, nutritional status, as determined by 
either the larval or adult environment, could influence foraging decisions. In addition, social facilitation 
(i.e., stimulation to feed by the presence of foraging conspecifics) 26 could in theory increase tolerance to 
a repellent. 

Biting pressure, also known as the “ambient biting rate,” is a condition basic to the measurement of 
repellent performance. This value may be measured in untreated subjects exposed to foraging 
ectoparasites. Higher biting pressures should correspond, in general, to greater parasite densities 
and, in nature, larger local population sizes and relatively fewer alternate sources of blood meals. 
Under high biting pressure conditions, repellents are likely to fail sooner because the encounter rate 
with the least sensitive foragers in the population will be great enough to cause failure based on 
absolute (e.g., first confirmed bite) rather than relative (percent biting reduction) criteria. 8 Similarly, 
efficacy tests with large numbers of subjects may sample more such insensitive mosquitoes, and 
perhaps even more on a per capita basis should group size enhance the detectability of hosts to 
parasites. Moreover, the availability of alternative host individuals may affect mosquito biting 
behavior and thus repellent performance. Repellents may be more effective when mosquitoes have 
the simultaneous option of choosing a more attractive host. 14 All of these basic factors should 
influence test design and conduct, but their importance may differ across mosquito species 
and conditions. 

Studies have also shown a number of more specific, user-mediated, proximate conditions that 
influence repellent performance. As is typical, most experimental data available are for deet 
formulations. Conditions of actual use that may reduce the duration of protection include contact with 
water, sweating, and abrasion by clothing or vegetation. ' Rueda et al. reached two main 
conclusions regarding the interaction of repellents and clothing. First, abrasion of treated skin by 
clothing fabric can significantly lower the protection afforded by a repellent. Second, the amount of 
friction between skin and fabric was increased by the presence of a repellent on the skin. This increase in 
friction likely aggravated the rate of its loss to the fabric. This study was conducted using the U.S. 
military polymer based extended duration deet formulation (EDTIAR). The generality of the results has 
not been explored with other formulations or active ingredients. 

Volatilization may be one of the most important variables, as it accounts for a major fraction of 
repellent loss from the skin. 30 In consequence, subject-caused differences in the rate of volatilization 
(whether related to physiological or activity differences) should be an important determinant of 
individual variation in repellent efficacy. However, no research appears to have directly examined the 
relationship between volatilization and repellency beyond the basic studies of Rutledge et al. 30 Costantini 
et al. 31 used the Rutledge method to model evaporation differences among repellents based on efficacy 
data, but they did not measure volatilization directly. Likewise, the extent to which conditions of use 
influence dermal absorption appears not to have been quantified. 


© 2006 by Taylor & Francis Group, LLC 



Evaluation of Topical Insect Repellents and Factors That Affect Their Performance 


251 


Formulation Chemistry 

Even within the standard test model of deet and Aedes aegypti, substantial variation in protection has 
been reported for decades. ' ’ " Given the many variables likely to underlie unexplained performance 
variation, Buescher et al. 33 reasoned that illuminating basic physical properties of repellent persistence 
could provide an important baseline for sensible repellent design. Using deet at a series of dilutions, they 
computed a dose-response curve describing the influence of concentration on the duration of 95% 
protection against caged Aedes aegypti. The curve is negatively exponential, meaning that each increase 
in concentration provides a progressively smaller increment in protection. Their main conclusion was 
that the Army 75% deet formulation achieved little added protection compared to, for example, a 50% 
concentration. This is a significant finding because use of lower concentrations would reduce deet’s 
plasticizing effects and toxicological risk values. 

While the importance of volatilization in limiting repellency duration was understood when the 
Buescher et al. 33 report appeared in the mid-1980s, it is likely that formalizing the dose-response 
relationship laid the foundation for a more analytical approach to designing extended-duration 
formulations that would deliver sufficient molecules for repellency over a predictable time span. 
Nonetheless, attempts to manipulate the chemistry of repellent carriers, whether through blending 
with a polymer or microencapsulation, to control volatilization (and dermal absorption at the same time) 
have met with mixed success. 

High volatility is likely to both enhance repellency and evaporation, leading to ephemeral protection. 
In the face of this tradeoff, Reifenrath and Rutledge 34 investigated the impact of numerous silicone 
polymers on the efficacy or protection time of deet against Aedes aegypti using dogs and mice. There 
was little influence in the dogs, and while 40% of the polymers increased performance in the mice, the 
changes were not large. Mehr et al. 35 examined controlled release polymers and starch microencapsula¬ 
tion of deet using the same mosquito species on white rabbits. Some increased duration of efficacy 
significantly, but none achieved better than 80% protection at 12 h. The efficacy results of a field test by 
Gupta et al. 36 that compared the Army 75% deet repellent with two candidate extended duration polymer 
formulations (Biotek with 42% deet and EDTIAR with 34% deet) are not interpretable for our purposes 
here, but important information on dosing did emerge. Ad libitum self-application resulted in an inverse 
relationship between deet concentration and the total amount of each formula applied, so that the mean 
quantity of deet applied differed little between the three products. 

This same inverse dosing relationship characterized a laboratory test of the same formulations against 
Aedes aegypti, Aedes taeniorhynchus, Anopheles Stephensi, and Anopheles albimanus by Gupta and 
Rutledge. 24 With a total of three subjects in three simulated climates. Biotek provided 94.9% protection, 
and EDTIAR 94.8% protection, from bites of all mosquito species in a series of exposures over 12 h. 
These values were superior to the 82% protection afforded by the Army 75% deet in ethanol. Enhanced 
performance in the polymerized formulations may stem from a combination of reduced volatilization and 
skin penetration. 37 Interestingly, Gupta and Rutledge 24 concluded that the EDTIAR was the best 
formulation because the performance of Biotek was “at best similar or less than that provided by the 3M 
formulation,” an assessment not consistent with the means they reported (above). In addition they cited 
the advantage of EDTIAR having the lowest deet concentration, but given the observed dosing (mean 
Biotek 0.9 mg/cm 2 , mean EDTIAR 1.1 mg/cm 2 ), more deet was actually delivered when the EDTIAR 
was applied. Overall, in spite of the excellent general design of this study, the use of only three study 
subjects limits the value of assessing the results at any greater level of detail or generalizing strongly 
from them. 

Two more recent studies, using laboratory rabbits and deet, have yielded clearer and more positive 
results concerning formulation and duration. Rutledge et al. 37 tested eight polymer and nine 
microencapsulated formulations. Against Aedes aegypti and Anopheles albimanus, several were more 
effective than unformulated deet at the same concentration for periods of up to 24 h. The best 
performance was with a polymer containing a high molecular weight fatty acid, and with micro- 
encapsulated formulations containing a diversity of large molecules, including lanolin, gums, acids, and 


2006 by Taylor & Francis Group, LLC 



252 


Insect Repellents: Principles, Methods, and Uses 


polypropylene glycol. In a study with argasid ticks, Salafsky et al. 38 reported that a liposomal 
formulation designed to reduce volatilization and dermal absorption extended the duration of repellent 
protection. In a three-day trial, attachment to a finger treated with liposomal deet was absent or 
significantly reduced compared to an equal concentration of deet in isopropanol, sampled at 24, 48, and 
72 h. Given the difficulty of preparing stable polymer formulations of deet, refined alternatives, including 
microcapsules and liposomes, should be considered for tests with other active ingredients and biting 
arthropods as well. 

Active Ingredients and Their Efficacy Assessment 

Active ingredients are the focus of most repellent development programs, and their efficacy is assessed 
through cage and field testing. The history of deet and other prominent repellents such as dimethyl 
phthalate is treated by Moore and Debboun in Chapter 1 and Strickman in Chapter 22. While it is 
accurate to state that a variety of subject factors and their interactions with other variables influence 
repellent performance, the review in the foregoing sections shows that the precise nature of those factors 
is poorly understood. At present, the chief manner in which the influence of such uncontrolled variation 
can be moderated (and studied) is by conducting tests with large numbers of subjects. 

This section reviews recent laboratory and field performance trials of promising non-deet repellents 
currently marketed in the U.S. and Europe. The goal is to apply insights gained from the foregoing review 
to evaluate how factors that influence repellent efficacy have been controlled and coordinated. Studies 
considered are mainly those treating Merck IR3535 (3-|/V-butyT/V-acetyl |-amino proprionic acid, ethyl 
ester), Lanxess Picaridin (aka KBR3023, (l-(l-methyl-propoxycarbonyl)-2-(2-hydroxy-ethyl)-piper- 
idine), and PMD (pora-menthane-3,8-diol, which is the prime repellent constituent of the U.S. EPA- 
registered active ingredient “oil of lemon eucalyptus,” from the tree Corymbia citriodora). These active 
ingredients were developed much more recently than deet; all are registered by the U.S. EPA. Most 
studies compare them to some type of deet standard. Given the variety of contingencies that apply to the 
performance of deet even under controlled conditions against well known mosquito taxa, it is worth 
examining how well conditions have been accounted for in tests of active ingredients that are less well 
studied. Frances (Chapter 18), Strickman (Chapter 20), and Puccetti (Chapter 21) also treat these three 
active ingredients in detail. 


Laboratory Efficacy Comparisons 

The most widely referenced recent study of comparative mosquito repellent efficacy was conducted with 
caged Aedes aegypti by Fradin and Day. 39 Their goal was to compare commercial deet products at 
various concentrations with plant-based repellents and IR3535 at 7.5%. Two lotions with at least 20% 
deet protected subjects for an average of 4-6 h (time to first bite), and most other formulations provided 
protection for well under 1 h. The authors concluded that “only products containing deet offer long- 
lasting protection.” The design was comparatively strong in terms of the number of test subjects (15), but 
the study had at least two apparent weaknesses. First, dosage was not reported and perhaps not closely 
controlled. Second, repellents that performed well in a subject’s first exposure were tested at less frequent 
intervals in the second and third exposures (apparently for convenience), adding a bias that probably 
exaggerated true differences among the products. Despite those shortcomings, the performance 
differences were large enough to suggest that conclusions were generally accurate. 

A substantially different picture emerged in the next broad-based cage study, 40 which included more 
effective commercial deet alternatives. Three mosquito species were tested separately: Culex nigri- 
palpus, Aedes albopictus, and Aedes triseriatus. Results for the four most effective products are 
highlighted in Table 12.2. Most remarkably, given deet’s five decade reign of superiority in such 
testing, overall repellency of the non-deet active ingredients was either consistently slightly greater (in 
the case of PMD), or equivalent to, 15% deet. However, for comparative purposes it is unfortunate that 
the highest deet concentration tested was only 15%. 


© 2006 by Taylor & Francis Group, LLC 


Evaluation of Topical Insect Repellents and Factors That Affect Their Performance 


253 


TABLE 12.2 


Mean Protection Time 3 (SE) (hours) for the Four Most Effective Repellents Studied in the Laboratory by 
Barnard and Xue 40 


Product 

Aedes albopictus 

Culex nigripalpus 

Aedes triseriatus 

Repel® (19.5% b PMD) 

7.8 (0.2) 

7.3 (0.7) 

7.8 (0.2) 

Bite Blocker® (2% soy oil) c 

5.5 (1.3) 

8.3 (0.2) 

7.8 (0.2) 

Autan® (10% Picaridin) 4 

5.7 (0.9) 

8.0 (0.0) 

7.8 (0.2) 

Off!® (15% deet) 4 

7.2 (0.8) 

7.0 (0.6) 

7.3 (0.3) 


a Time to second bite in one or two sequential periods. 

b Corrected from Barnard and Xue 40 ; a registered trademark of Wisconsin Pharmacol Co.. Inc., Jackson, WI. 
c Methylated soy bean oil; a registered trademark of HOMS, LLC, Clayton, NC. 

4 Registered trademarks of S.C. Johnson and Son, Inc., Washington, DC. 

Source: From D. R. Barnard, and R. D. Xue, Journal of Medical Entomology, 41(4), 726-730, 2004. 


Strengths of that study include that the repellents were applied at a standard dosage (1 mL/650 cm 2 of 
skin surface), and tested against a high density of avid mosquitoes. However, an important weakness was 
that only two subjects tested each repellent, out of a total of five subjects. Because individuals differ 
inherently in their attractiveness to mosquitoes and dermal interaction with repellents, and both factors 
interact with mosquito taxon, a substantial portion of the variation reported may be from uncontrolled 
subject error. 

Cage studies against Anopheles vectors of Plasmodium (malaria) likewise showed PMD 41-43 and 
Picaridin 15 to be at least as effective as deet formulations. The first three tests had six or fewer subjects 
and uncontrolled or unspecified dosing. 41 Badolo et al. 15 also found Picaridin to be more effective than 
deet against an African strain of Aedes aegypti , but the number of subjects and biting pressure were not 
reported. Data in Carroll and Loye 44 suggested that 19.5% PMD was intermediate in performance 
between ten and 30% deet products against Aedes aegypti over an eight hour period (eight PMD subjects, 
one subject for each deet formulation, with equivalent dosing and biting pressure of 50 bites/min on 
untreated arms). All of these studies would benefit from larger samples or more complete reporting. One 
major benefit from more replication would be more realistic comparisons between separate studies. 

There have been fewer studies of IR3535 at higher concentrations than the basic 7.5% Avon formula 
(above), but there is an indication that efficacy improves. At 20% IR3535, a study of three subjects at 
high biting pressures by Thavara et al. 45 found IR3535 comparable to 20% deet against two Culex and 
one Aedes species, but less repellent against an Anopheles species. 

Field Efficacy Comparisons 

Most field efficacy trials share problems common in laboratory trials, including small numbers of 
subjects, lack of repetition, uncontrolled dosing, and unclear ambient biting rates. As a result, 
characterizing the repellency of a given active ingredient across taxa, and comparing it with other 
active ingredients, is difficult to do at a suitable level of precision. 

One of the most thorough and thoughtful studies of contemporary active ingredients was conducted by 
Costantini et al., 31 measuring dose-response curves of deet, Picaridin and IR3535 against Anopheles 
gambiae complex mosquitoes in Burkina Faso. Eight male subjects tested a series of dosages of the 
technical grade repellents diluted in ethanol. Apparently each repellent was tested on 96 nights (12 times by 
each subject). Testing was performed over the ten hour period 18:00-04:00 with a two hour break from 
22:00-00:00. Four dosages (in ethanol) were tested, specifically 0.1, 0.3, 0.6, and 0.8 mg/cm 2 of each 
active ingredient. For comparison, standard volume for efficacy testing in the U.S. is ca. 1.54 mg/cm 2 , so 
that for a 30% (high) concentration active ingredient, dosing would equal about 0.5 mg/cm 2 of active 
ingredient. The two higher doses in this study were thus greater than those intended for most military or 


2006 by Taylor & Francis Group, LLC 



254 


Insect Repellents: Principles, Methods, and Uses 


popular commercial formulations in the U.S. Picaridin performed best against the anophelines in this 
study, with an estimated 95% or more repelled for at least 8 h at the three higher dosages. Deet's 
performance was intermediate, and IR3535 was the least repellent at all dosages. These results are 
important because even though deet is historically the best repellent against anophelines, public health 
professionals have long recognized the need for a better repellent against these important vectors of the 
pathogens causing malaria. The 0.3 mg/cm 2 dosage corresponded to a 20% Picaridin formulation, the 
maximum concentration that is registered for use in Europe and Australia. Costantini et al. 31 provide some 
of the first evidence of a repellent lasting for such a long period against Anopheles gambiae (see also 
Trigg 46 below for PMD performance). 

As in other studies, however, caution is in order. First, in spite of the unusually long duration of the 
study (six months in total), which yielded an unusually large data set, just eight subjects were involved, 
and only local populations of Anopheles gambiae. Second, although samples for other mosquito taxa 
were small, Picaridin did not repel Aedes species better than the other repellents. Third, while control 
subjects collected a large number (27,231) of alighting Anopheles gambiae during the study, arithmetic 
shows this to be a low ambient biting rate for the study: less than 0.3 per minute (27,231 
bites/92,160 min). For perspective, current U.S. EPA guidelines call for a minimum of 1 bite/min on 
a lower limb (feet and hands excluded), more than three times greater than the observed rate. So while 
the strength of this study is that it was conducted under representative (long-term) conditions, and low 
biting rates may be medically important when infection rates in mosquitoes are high, it would still be 
valuable to have performance data at higher biting rates. Lastly, data from women are clearly merited. 

Even at such low biting rates, Picaridin may fail quickly against anophelines. Frances et al. 50 tested 
19.2% Picaridin (Autan Repel Army 20) against 20% deet in ethanol and 35% deet in a gel (the repellent 
issued by the Australian Defense Force) against Anopheles meraukensis and Anopheles bancroftii in 
Australia’s Northern Territory. At control biting rates slightly under 0.5 bites/min, 35% deet and 
Picaridin protected at more than 95% over the first hour, but by the second hour repellency dropped to 
78% for Picaridin, and declined variably in all three repellents thereafter. Those data were collected by 
four subjects, all male, with each testing a repellent or ethanol control twice over eight consecutive 
nights. Dosage appears to have been ad libitum, determined by the subjects at the time of application. By 
weight, one can calculate that Picaridin was applied at an average rate 31% higher than the 20% deet, and 
45% higher than the 35% deet. In this latter case, only about 25% more deet than Picaridin was actually 
administered (estimated from Table 12.1 of Frances et al. 50 ). The rate at which formulated Picaridin was 
applied averaged 13% higher than standard procedure for a U.S. repellent efficacy test (1 mL/650 cm 2 of 
skin surface). 

Other field tests of Picaridin against anophelines are similarly plagued by small samples or low 
ambient biting rates (<0.5/min, e.g.. Yap et al. 47 ' 48 ), but still suggest its promise as a broad-spectrum 
mosquito repellent. In the single test conducted at high ambient biting rates, Barnard et al. 49 compared 
25% ethanol solutions of technical deet and IR3535, and Picaridin, and PMD at 19.5% in a commercial 
lotion (not 40% PMD as indicated in the source publication; see Carroll and Loye 44 ). Five males exposed 
treated limbs for 3 min each hour for 6 h, beginning 15 min after application. The test was repeated five 
times over three days so that each subject tested each repellent and served as a control (25% water in 
ethanol) once. Black salt marsh mosquitoes ( Aedes taeniorhynchus) attacked control subjects at a high 
average rate of 19.5 + 13.7 bites/min. Given the small number of subjects, statistical power was low, but 
Picaridin and deet appeared to be the most repellent, followed by PMD and then IR3535. Only Picaridin 
repelled at greater than 95% through hour five. 

The efficacy of PMD against anophelines appears noteworthy. Using six self-dosed subjects exposed 
to Anopheles gambiae in rural Tanzania, Trigg 46 compared 50% PMD to 50% deet under low ambient 
biting rate conditions (apparently 0.13/min, calculated from grand mean of controls over the 240 min 
exposure period, Trigg’s Table 1). Repellents were applied 5h before the onset of exposure. Deet 
prevented all biting on six subjects for close to 7 h, and PMD for 6-8 h, depending on formulation. 
Moore et al. 50 collected similar data for Anopheles darlingi in Bolivia, but tested only 2^1 h after 
application. PMD (30%) reduced biting on five subjects by a mean of 97%, while 15% deet in ethanol 


© 2006 by Taylor & Francis Group, LLC 


Evaluation of Topical Insect Repellents and Factors That Affect Their Performance 


255 


gave just 85% protection. Compared to other studies of anophelines, ambient biting pressure was 
respectably high, greater than or equal to 1 bite/min (estimated from the mean percentage biting rate 
reductions of the test products, including 0% for the control, and the total number of mosquitoes captured 
landing). Variation in the performance of Picaridin among anophelines (e.g., Frances et al. 51 above) 
suggests that PMD, too should be tested against more anopheline species, using controlled dosing on 
more study subjects than in the foregoing studies. 

In a six hour held study of PMD with a large number (20) of adult male and female subjects exposed to 
Aedes melanimon and Aedes vexans in California, Carroll and Loye 44 found excellent protection with 
continuous exposure of lower arms and legs at mean biting pressures of approximately 1.5 and 3 per 
minute, respectively. Subjects tested lotion (19.5% PMD) and spray (26% PMD) formulations at dosages 
of either 1.6 or 2.4 mg/cm 2 . Mean biting rate reduction for all treatments over the 6h was 99.9%. 
Protection provided by 20% deet lotion was similar, but only two subjects tested deet. 

Other than Barnard et al. 49 held studies of IR3535 at higher concentrations are rare. Thavara et al. 45 
compared IR3535 and deet at a rate of 20% in ethanol with six subjects against several mosquito species at 
low biting rates. In two 8 h held studies of Aedes albopictus at ambient biting rates of about 0.35 bites/ 
min, there were no bites from this species on subjects using either repellent. IR3535 reduced biting by a 
mean of 98.4%. Deet reduced biting by 97.4%. The authors’ claim that the difference, statistically 
signihcant at P<0.05, is inconsequential, however, given the similarity of the means (see Table 12.1 of 
referenced study). Protection in similar hve hour studies against night-biting Culex, Mansonia and several 
Anopheles species (ambient biting pressure 0.15-0.25 in the last genus) averaged 98% and greater for 
both repellents. 45 Doses were approximately double the standard. Like studies of other promising 
repellents, work on IR3535 would beneht from greater standardization of protocols, more subjects, and 
higher biting rates. 


Conclusions 

The task of generating predictable, generalized results from insect repellent efficacy tests is challenging. 
The basic difficulty is in the effort to generate and deliver chemicals that will interrupt the feeding 
behavior of highly diverse and refined biting arthropods without harming the user. Even among 
apparently safe and effective repellent candidates, however, this review demonstrates that the interplay 
of host, arthropod, environment, and utility significantly controls performance. We have a better idea of 
what classes of variables are influential than we do of how to predict the impact of a particular variable in 
any given case. 

For each active ingredient, the basic three-way interplay between subject, formulation, and mosquito 
taxon appears to be the principal source of variation in the outcomes of efficacy tests. It is typically an 
uncontrolled source of error that hinders all attempts to analyze additional conditional factors (e.g., 
environment, use). Because this axis of interaction is poorly described, the precision of performance 
estimates is generally questionable. 

For practical reasons, most studies have attempted to assess variables one or two at a time. From these 
we can begin to list factors that should be included in improved models of repellent performance. For 
example, it is likely that variation in skin emanations, including lipids, 22 moisture and heat 19 affect 
attractiveness, and that variation in dermal absorbency (three-fold for deet 23 ) affects protection time. 
These results can be linked to build a nascent picture of subject variation. At the same time, however, two 
large-scale studies of gender that both reported highly significant and substantial effects produced 
strikingly dissimilar results: Gilbert, Gouch, and Smith 19 found that females were clearly less susceptible 
to biting by Aedes aegypti in a test of 5% deet in alcohol, while Golenda et al. 20 found that females were 
clearly more susceptible to biting by Anopheles stephensi in a test of EDTIAR. So while we can state that 


© 2006 by Taylor & Francis Group, LLC 




256 


Insect Repellents: Principles , Methods, and Uses 


gender is an important consideration for repellent performance, at present, uninvestigated interactions 
between gender, repellent formulation, and mosquito taxon prevent us from offering further direction. 

The problems of inconsistency in the design, execution, and reporting of efficacy studies likewise 
hinder the effort to evaluate repellent performance. In all of the studies reviewed that compare active 
ingredients, sample sizes are too small to permit confident distinction among treatments with moderately 
close performance values. Even when statistical significance is shown, differences cannot necessarily be 
attributed to the repellents alone, and peculiarities of individual interactions may be paramount. Note, for 
example, that many studies have used only male subjects. Dosage is another factor of obvious 
importance 33 that is too often uncontrolled. 

The use of limited numbers of subjects to test repellents probably has its justifications in the desire to 
minimize risk, the difficulties associated with recruiting people for this type of work, the use of the first 
confirmed bite criterion (a threshold measure), and perhaps also in the history of testing deet, for which 
relative variation was apparently regarded as inconsequential due to its outstanding comparative efficacy. 
This tradition is reflected in guidelines for efficacy testing proposed by the U.S. EPA, requiring just six 
subjects for the generation of registration data. We have entered a new era in which there is for the first 
time an interest in comparing several repellent active ingredients, all of high efficacy. How shall they be 
distinguished? 

Rutledge and Gupta 52 determined by meta-analysis of published studies that the standard deviations of 
protection times are a linear function of the means. As a result, the statistical differentiation of long-acting 
formulations in particular will probably require especially large samples. It is unfortunate that n = 20, the 
minimum acceptable sample size for parametric hypothesis testing at alpha-levels of 0.05, is not the norm 
for repellent studies. While the Rutledge and Gupta 52 estimate of required n’s is likely inflated by 
interstudy variation beyond that relevant to any given comparison (e.g., of two formulations tested 
simultaneously), their study does give the impression that even 20 subjects per formula might be too few. 
Nonetheless, an agenda to deploy large, balanced groups of subjects to test various repellents against 
various mosquitoes would likely advance repellent science substantially. 

The complementary perspective is to accept that separating the performances of candidate deet- 
replacements is a futile exercise. A positive outcome of that view might be to open the door more readily 
to inclusive strategies, such as combining active ingredients to see if, for example, variance in 
performance can be limited. Reducing variance is important because we tend to rely on the mean 
protection period when evaluating performance. Any subgroup of people that is less protected than 
average will be systematically less protected than is otherwise assumed. 

In addition, it is sensible to make inferences from the results of many different studies that involve the 
same repellents. For example, Picaridin seems to be especially efficacious in many of the studies 
reviewed here. Meta-analyses of such data sets potentially have the added advantage of treating data 
from tests conducted in a variety of conditions against a variety of mosquito taxa. At the same time, the 
present suite of studies available seems to share inherent biases (male subjects) and serious 
inconsistencies (very unequal dosing) that obscure their value to objective analysis. If we coordinate 
and more thoroughly standardize the conduct and reporting of repellent studies in the future, interested 
scientists, health professionals, and the public will all benefit from the resulting increase in 
available knowledge. 

Acknowledgments 

For guidance and assistance at UC-Davis, I thank R. Washino, B. Eldridge, T. Scott, J. Loye, S. Lawler, 
and S. Minnick. Additional insight has come from D. Barnard, W. Reifenrath, L. Rutledge, M. Schneider, 
W. Wakesa, G. White, M. Wundrock, and the editors of this book. 

References 

1. R. K. Gupta and L. C. Rutledge, Role of repellents in vector control and disease prevention. American 
Journal of Tropical Medicine and Hygiene, 50, 82-86, 1994. 


© 2006 by Taylor & Francis Group, LLC 



Evaluation of Topical Insect Repellents and Factors That Affect Their Performance 


257 


2. M. Debboun and D. Strickman, Repellents and the military: Our first line of defense. Wing Beats, 14, 
30-31, 2003, see also pages 33-34. 

3. U.S. Environmental Protection Agency. Office of Pesticides and Toxic Substances, Special Pesticide 
Review Division, iV,/V-diethyl-m-toluamide (DEET) Pesticide Registration Standard (EPA-540/RS- 
81-004), U.S. Environmental Protection Agency, (PB81-207722), Washington, D.C., 1980. 

4. E. E. Davis, Insect repellents: Concepts of their mode of action relative to potential sensory 
mechanisms in mosquitoes ( Diptera: Culicidae ), Jounal of Medical Entomology, 22, 237, 1985. 

5. W. A. Skinner and H. L. Johnson. The design of insect repellents, in Drug Design, E. J. Ariens (Ed.), 
New York: Academic Press, 1980, pp. 277-305. 

6. W. V. King, Chemicals Evaluated as Insecticides and Repellents at Orlando, Florida, Entomology 
Research Branch, Department of Agriculture, Washington, DC, 1954. 

7. C. N. Smith, I. H. Gilbert, H. K. Gouck, M. C. Bowman, F. Acree, and C. H. Bowman, Factors 
affecting the protection periods of mosquito repellents. United States Department of Agriculture 
Technical Bulletin, 1285, 1-36, 1963. 

8. L. C. Rutledge, M. A. Moussa, C. A. Lowe, and R. K. Solfield, Comparative sensitivity of mosquito 
species and strains to the repellent diethyl toluamide. Journal of Medical Entomology, 14, 536-541, 
1978. 

9. P. Granett, Studies of mosquito repellents. I. Test procedure and method for evaluating data. Journal of 
Economic Entomology, 33, 563-565, 1940. 

10. A. A. Khan, H. I. Maibach, and D. L. Skidmore, Insect repellents: Effect of mosquito and repellent- 
related factors on protection time. Journal of Economic Entomology, 68, 43-45, 1975. 

11. D. R. Barnard, K. H. Posey, D. Smith, and C. E. Schreck, Mosquito density, biting rate and cage size 
effects on repellent tests, Medical and Veterinary Entomology, 12, 39-45, 1998. 

12. B. V. Travis, Relative effectiveness of various repellents against Anopheles farauti Laveran, Journal 
of the National Malaria Society, 6, 180-183, 1947. 

13. L. C. Rutledge, D. M. Collister, V. E. Meixsell, and G. H. G. Eisenberg, Comparative sensitivity of 
representative mosquitoes ( Diptera: Culicidae ) to repellents, Journal of Medical Entomology, 20, 
506-510, 1983. 

14. C. F. Curtis, J. D. Lines, J. Ijumba, A. Callaghan, N. Hill, and M. A. Karimzad, The relative efficacy of 
repellents against mosquito vectors of disease, Medical and Veterinaiy Entomology, 1, 109-119, 1987. 

15. A. Badolo, E. Ilboudo-Sanogo, A. P. Ouedraogo, and C. Costantini, Evaluation of the sensitivity of 
Aedes aegypti and Anopheles gambiae complex mosquitoes to two insect repellents: DEET and KBR 
3023, Tropical Medicine & International Health, 9(3), 330-334, 2004. 

16. R. E. Coleman, L. L. Robert, L. W. Roberts, J. A. Glass, D. C. Seeley. A. Laughinghouse, P. V. Perkins, 
and R. A. Wirtz, Laboratory evaluation of repellents against four anopheline mosquitoes 
( Diptera: Culicidae ) and two phlebotomine sand flies ( Diptera: Psychodidae), Journal of Medical 
Entomology, 30, 499-502, 1993. 

17. R. A. Wirtz, E. D. Rowton, J. A. Hallam, P. V. Perkins, and L. C. Rutledge, Laboratory testing of 
repellents against the sand fly Phlebotomus papatasi ( Diptera: Psychodidae), Journal of Medical 
Entomology, 23, 64-67, 1986. 

18. L. C. Rutledge, R. K. Gupta, G. N. Piper, and C. A. Lowe, Studies on the inheritance of 
repellent tolerance in Aedes aegypti, Journal of the American Mosquito Control Association, 10, 
93-100, 1994. 

19. I. H. Gilbert, H. K. Gouck, and N. Smith, Attractiveness of men and women to Aedes aegypti and relative 
protection time obtained with deet, Florida Entomologist, 49, 53-66, 1966. 

20. C. F. Golenda, V. B. Solberg, R. B. Burge, J. M. Gambel, and R. A. Wirtz, Gender-related efficacy 
difference to an extended duration formulation of topical AGV-diethyl-m-toluamide (deet), American 
Journal of Tropical Medicine and Hygiene, 60, 654-657, 1999. 

21. R. L. Reitschel and P. S. Spencer, Correlation between mosquito repellent protection time and insensible 
water loss from the skin. Journal of Investigative Dermatology, 65, 385-387, 1975. 

22. W. A. Skinner, H. C. Tong, H. Johnson, R. M. Parkhurst, D. Thomas, T. Spencer, W. Acres, 
D. Skidmore, and H. Maibach, Influence of human skin surface lipids on protection time of topical 
mosquito repellent, Journal of Pharmaceutical Science, 66, 1764-1766, 1977. 


2006 by Taylor & Francis Group, LLC 



258 


Insect Repellents: Principles, Methods, and Uses 


23. S. Selim, R. E. Hartnagel, T. G. Osimitz, K. L. Gabriel, andG. P. Schoenig, Absorption, metabolism, and 
excretion of A,,/V-diethyl-m-toluamide following dermal application to human volunteers, Fundamental 
and Applied Toxicology, 25, 95-100, 1995. 

24. R. K. Gupta and L. C. Rutledge, Controlled release repellent formulations on human volunteers under 
three climatic regimens. Journal of the American Mosquito Control Association, 7, 490-493, 1991. 

25. D. R. Barnard. Mediation of deet repellency in mosquitoes (Diptera: Culicidae) by species, age, and 
parity. Journal of Medical Entomology, 35(3), 340-343, 1998. 

26. A. Ahmadi and G. H. A. McLelland, Mosquito-mediated attraction of female mosquitoes to a host. 
Physiological Entomology, 10, 251-255, 1985. 

27. L. C. Rutledge, W. G. Reifenrath, and R. K. Gupta. Sustained release formulations of the U.S. Army 
insect repellent. Proceedings of the Army Sciences Conference, 3, 343-357, 1986. 

28. V. B. Solberg, T. A. Klein, K. R. McPherson, B. Bradford, J. R. Burge, and R. A. Wirtz, Field evaluation 
of deet and a piperidine repellent (A13-37220) against Amblyomma americanum ( Acari: Ixodidae), 
Journal of Medical Entomology, 32, 870-875, 1995. 

29. L. M. Rueda, L. C. Rutledge, and R. K. Gupta, Effect of skin abrasions on the efficacy of the repellent 
deet against Aedes aegypti. Journal of the American Mosquito Control Association, 14, 178-182,1998. 

30. L. C. Rutledge, R. A. Wirtz, M. D. Buesher, and Z. A. Mehr, Mathematical models of the effectiveness 
and persistence of mosquito repellents. Journal of the American Mosquito Control Association, 1, 
56-62, 1985. 

31. C. Costantini, A. Badolo, and E. Ilboudo-Sanogo, Field evaluation of the efficacy and persistence of 
insect repellents DEET, IR3535, and KBR 3023 against Anopheles gambiae complex and other 
Afrotropical vector mosquitoes. Transactions of the Royal Society of Tropica! Medicine and 
Hygiene, 98(11), 644-652, 2004. 

32. C. E. Schreck, Techniques for the evaluation of insect repellents: A critical review. Annual Review of 
Entomology, 22, 101-119, 1977. 

33. M. D. Buesher, L. C. Rutledge, R. A. Wirtz, and J. H. Nelson, The dose-persistence relationship of deet 
against Aedes aegypti. Mosquito News, 43, 364-366, 1983. 

34. W. G. Reifenrath and L. C. Rutledge, Evaluation of mosquito repellent formulations. Journal of 
Pharmaceutical Science, 73, 169-173, 1983. 

35. Z. A. Mehr, L. C. Rutledge, E. L. Morales, V. E. Meixsell, and D. W. Corte, Laboratory evaluation of 
controlled-release insect repellent formulations. Journal of the American Mosquito Control Associ¬ 
ation, 1, 143-147, 1985. 

36. R. K. Gupta, A. W. Sweeney, L. C. Rutledge, R. D. Cooper, S. P. Frances, and D. R. Westrom, 
Effectiveness of controlled-release personal-use arthropod repellents and permethrin impreg¬ 
nated clothing in the field. Journal of the American Mosquito Control Association, 3, 556-560, 1987. 

37. L. C. Rutledge, R. K. Gupta, Z. A. Mehr, M. D. Buesher, and W. G. Reifenrath, The evaluation of 
controlled-release mosquito repellent formulations. Journal of the American Mosquito Control 
Association, 12, 39-44, 1996. 

38. B. Salafsky, Y. He, J. Li, T. Shibuya, and K. Ramaswami, Short report: Study on the efficacy of a new 
long-acting formulation of WA-diethyl-m-toluamide (deet) for the prevention of tick attachment, 
American Journal of Tropical Medicine and Hygiene, 62, 169-172, 2000. 

39. M. S. Fradin and J. F. Day, Comparative efficacy of insect repellents against mosquito bites. New 
England Journal of Medicine, 347, 13-18, 2002. 

40. D. R. Barnard and R. D. Xue, Laboratory evaluation of mosquito repellents against Aedes albopictus, 
Culex nigripalpus, and Ochlerotatus triseriatus (Diptera: Culicidae), Journal of Medical Entomology, 
41(4), 726-730, 2004. 

41. J. K. Trigg and N. Hill, Laboratory evaluation of a eucalyptus-based repellent against four biting 
arthropods , Phytotherapy Research, 10, 313-316, 1996. 

42. N. Hill, Insect repellent test report, unpublished document, London School of Hygiene and Tropical 
Medicine, 1998. 

43. J. Govere, D. N. Durrheim, L. Baker, R. Hunt, and M. Coetze, Efficacy of three insect repellents 
against the malaria vector Anopheles arabiensis. Medical and Veterinary Entomology, 14, 
441^144, 2000. 


© 2006 by Taylor & Francis Group, LLC 



Evaluation of Topical Insect Repellents and Factors That Affect Their Performance 


259 


44. S. P. Carroll and J. E. Loye, A commercially-available botanical insect repellent as effective as 
deet, Journal of the American Mosquito Control Association, in press. 

45. U. Thavara, A. Tawatsin, J. Chompoosri, W. Suwonkerd, U. Chansang, and 
P. Asacadachanukorn, Laboratory and field evaulations of the insect repellent IR3535 (ethylbu- 
tylacetylaminopropionate) and deet against mosquito vectors in Thailand, Journal of the American 
Mosquito Control Association, 17, 190-195, 2001. 

46. J. K. Trigg, Evaluation of a eucalyptus-based repellent against Anopheles spp. in Tanzania, 
Journal of the American Mosquito Control Association, 12, 243-246, 1996. 

47. H. H. Yap, K. Jahangir, A. S. C. Chong, C. R. Adanan, N. L. Chong, Y. A. Malik, and 

B. Rohaizat, Field efficacy of a new repellent, KBR 3023, against Aedes albopictus (SKUSE) and 

Culex quinquefasciatus (SAY) in a tropical environment, Journal of Vector Ecology, 23, 62-68, 
1998. 

48. El. H. Yap, K. Janangir, and J. Zairi, Field efficacy of four insect repellent products against vector 
mosquitoes in a tropical environment, Journal of the American Mosquito Control Association, 16, 
241-244, 2000. 

49. D. R. Barnard, U. R. Bernier, K. H. Posey, and R. D. Xue, Repellency of IR3535, KBR3023, 
pflra-menthane-3,8-diol, and deet to black salt marsh mosquitoes ( Diptera: Culicidae) in the 
Everglades National Park, Journal of Medical Entomology, 39(6), 895-899, 2002. 

50. S. J. Moore, A. Lenglet, and N. Hill, Field evaluation of three plant-based insect repellents 

against malaria vectors in Vaca Diez province, the Bolivian Amazon, Journal of the American 

Mosquito Control Association, 18, 107-110, 2002. 

51. S. P. Frances, N. V. Dung, N. W. Beebe, and M. Debboun, Field evaluation of repellent 
formulations against daytime and nighttime biting mosquitoes in a tropical rainforest in northern 
Australia, Journal of Medical Entomology, 39, 541-544, 2002. 

52. L. C. Rutledge and R. K. Gupta, Variation in the protection periods of repellents on individual 
human subjects: An analytical review. Journal of the American Mosquito Control Association, 15, 
348-355, 1999. 


2006 by Taylor & Francis Group, LLC 



© 2006 by Taylor & Francis Group, LLC 



13 


Repellents Used in Fabric: The Experience 
of the U.S. Military 


Wilfred C. McCain and Glenn J. Leach 


CONTENTS 

Introduction.261 

Repellents and Fabric Treatment.262 

Repellent Application to Fabric.265 

Fland Application.266 

Barrier Method.266 

Spray Method.267 

Dust Method.267 

Immersion Method.267 

Factory Pretreatment.267 

Long-Lasting Surface Treatment of Fabrics with Insecticides.268 

Olyset® Technology and Production.269 

Industrial Production of Long-Lasting Insecticide-Treated Yarns.269 

Safety of Clothing Treatments.269 

References.271 


Introduction 

Humans are the only mammals that routinely wear clothing. The obvious functional comparison is 
between our clothing and the pelage of mammals or birds. Just as our use of clothing varies with our 
intended activity, the nature of fur or feathers can vary with specific adaptation of a species and from 
season to season for a single species. Considering this comparison, it is perhaps not surprising that many 
of the arthropods that bite humans interact with clothing. Chigger mites and tropical rat mites tend to bite 
where clothing is most closely pressed to the skin. Mosquitoes and stable flies commonly bite right 
through clothing, especially when the weave creates openings and the cloth lays flat on the skin. Body 
lice actually require clothing for resting and oviposition sites. Complete systems for personal protection 
from biting arthropods must consider the means to enhance clothing as a barrier, as well as protect 
exposed skin by the use of topical repellent products. 


261 


© 2006 by Taylor & Francis Group, LLC 





















262 


Insect Repellents: Principles , Methods, and Uses 


Repellents and Fabric Treatment 

In May of 1942, Philip Granett, working under a National Carbon Company Fellowship at Rutgers 
University, found that a cheesecloth sleeve impregnated with the repellent butyl carbitol acetate 
protected an untreated arm against mosquitoes for 24 h. In progress reports he sent to the company, 
he stated that some materials were effective for longer periods of time on cloth than on skin. Preliminary 
laboratory and field tests conducted in Florida in 1942 demonstrated that mosquitoes avoided some 
treated fabrics for several days; whereas, repellents applied to the skin failed to give complete protection 
after a few hours. 1 This finding started a new era in the field of personal protection from arthropods of 
public health importance. Madden and Lindquist 2 in 1944 and by McCullough and Jones 3 in 1945 
summarized studies showing the relative effectiveness of several repellents when applied to clothing. 

Prior to World War II, reports from the Panama Canal and other tropical areas prompted the 
Quartermaster General’s Office to review the entire problem of “adaptation of clothing to the physiologic 
requirements of the soldier.’’ 4 In 1941, The National Research Council approved support for the Harvard 
Fatigue Laboratory to investigate “clothing, fatigue, and supplementary substances.” 

Chemical substances that were to be impregnated into clothing materials to protect against chemical 
warfare agents, fire, and insects had to be effective, nonirritating and nontoxic. The Surgeon General 
tasked the Orlando Laboratory of the United States Department of Agriculture (USDA) to develop 
methods to control arthropods of medical interest. This resulted in a massive effort to quickly identify and 
appraise the potential of chemical compounds and mixtures. The USDA in conjunction with a number of 
other agencies including the U.S. Army, U.S. Navy, Rockefeller Foundation, Tennessee Valley 
Authority, the U.S. Public Health Service, and various groups working under contract to the Office of 
Scientific Research and Development entered into this effort. Collaborative research with other allied 
nations was also conducted and a vast amount of information was exchanged. Industry submitted most of 
the chemicals tested for their effectiveness against a variety of insects and arthropods. 

During World War II, scientists evaluated more than 10,000 such materials. By 1950, more than 
20,000 chemicals and mixtures had been tested for efficacy against arthropods and more than 6,000 had 
been tested as clothing treatments. 5 Of these, only a handful met the criteria of being both effective and 
safe. The Army selected some of these for general use. 

During this period, the USDA had scientists working under a fellowship that Rutgers University 
organized with Philip Granett in charge. The fellowship funded the development of an economic 
mosquito repellent that would be effective, safe, and acceptable for application to the skin. This 
collaboration resulted in the development of Insect Repellent 612 (2-ethyl-1,3-hexanediol) that was 
seven times more effective than citronella and superior to it in other respects. The repellent properties of 
another compound, dimethyl phthalate, had been discovered by another Rutgers Fellowship sponsored 
by the Standard Oil Company of New Jersey in the late 1930s and a third compound, indalone (2,2- 
dimethyl-6-carbobutoxy-2,3-dihydro-4-pyrone), was originally tested by the fellowship in 1937. 
Researchers found these agents effective against mites as well as several other types of arthropods. 6 

Extensive testing both in the laboratory and in the field followed the development of dimethyl 
phthalate, indalone, Rutgers 612, as well as dibutyl phthalate. Madden and his colleagues 2 reported the 
findings of field studies conducted in Orlando in 1942, using 28 different materials or combination of 
materials including 10 kinds of dusts, 1 ointment, and 3 liquid formulations as skin and clothing 
repellents. Skin testing was discontinued when it was demonstrated that liquid repellents applied to 
clothing were far more effective for protection from mites, considered key vectors at the time because of 
the deadly scourge of scrub typhus in the Pacific Theater. Scientists found indalone, Rutgers 612, and 
dimethyl phthalate all effective in protecting individuals against mites. The application of any of these 
three compounds in solvents to fabrics gave protection for up to 39 d and undiluted dimethyl phthalate 
protected people for 59 d, especially when applied around the openings of clothing. This led to the army’s 
adoption of dimethyl phthalate to control scrub typhus. It was applied as a spray to the clothing or used as 


© 2006 by Taylor & Francis Group, LLC 



Repellents Used in Fabric: The Experience of the U.S. Military 


263 


an emulsion in which clothing was dipped and then dried. Bulk supply of dimethyl phthalate in gallon 
containers for application to clothing was recommended in the fall of 1943 for certain of the overseas 
theaters. In December 1944, it was recommended that an emulsifier be incorporated with the dimethyl 
phthalate to facilitate the preparation of emulsions in the field. 7 In 1945, members of the United States of 
America Typhus Commission working in New Guinea developed a field method for the impregnation of 
clothing with a 5% emulsion of dimethyl phthalate in soapy water. This treatment was used in endemic 
areas for protection of both combat and staging troops. Research found this more practicable than 
previously used methods of applying repellents to clothing and was widely employed as a preventive 
measure to protect troops from scrub typhus in the western Pacific. 

Because chemicals to manufacture dimethyl phthalate were in short supply during World War II, 
benzyl benzoate was used as a miticide and eventually became the standard by which researchers 
evaluated future clothing repellents. As investigations of repellents continued, scientists discovered a 
number of substances that were superior to dimethyl phthalate for impregnation of clothing to protect 
against larval mites. They selected benzyl benzoate because of its rapid action against mites and its 
persistence in clothing after laundering. It was used in various mixtures for impregnation of clothing to 
repel mosquitoes, gnats, ticks and flies). 

In March 1945, a recommendation was made to the Office of the Quartermaster General that benzyl 
benzoate, together with an emulsifier, be substituted for dimethyl phthalate in the bulk issue of insect 
repellent. Because of difficulties in procurement, supplies of benzyl benzoate did not reach the field in 
time to be of use before the end of the war. It was used through the action in Vietnam to control and repel 
certain ticks and mites 8 and was available for clothing application as 90% benzyl benzoate. 

With the exception of occasional cases of skin irritation, few adverse effects have been reported from 
the use of benzyl benzoate. The efficiency of skin absorption is not known. Absorbed benzyl benzoate is 
rapidly biotransformed to hippuric acid that is excreted in the urine. When given in large doses to 
laboratory animals, benzyl benzoate causes excitement, lack of coordination, paralysis of the limbs, 
convulsions, respiratory paralysis, and death. No human poisonings have been reported. 9 The U.S. 
Environmental Protection Agency (EPA) is currently reviewing data submitted by the producers 
regarding benzyl benzoate’s human health and environmental effects. This data will be used to determine 
the pesticide’s eligibility for reregistration and will result in a Reregistration Eligibility Decision (RED) 
document. 10 

To develop a repellent with efficacy against a greater variety of arthropods, scientists used a 
combination of repellents. Formula 6-2-2 (M-250), a mixture containing 60% dimethyl phthalate, 
20% Rutgers 612, and 20% indalone, proved to be a highly effective repellent. The 6-2-2 formula was 
tested as a clothing treatment and compared with both dimethyl phthalate and Rutgers 612 in 1942- 
1943.' The testing showed that there was virtually no difference in protection among the three 
products tested. 

Another mixture of several effective repellents was developed specifically as a clothing repellent at the 
Orlando laboratory around this time and given the code number M-1960. This mixture contained 30% 
each of 2-butyl-2-ethyl-1,3-propanediol, benzyl benzoate, (V-butylacetanilide and 10% of a surfactant, 
Tween 80. This combination proved to be effective against mosquitoes, biting flies, fleas, mites, and ticks 
when applied to clothing at 3.9 mg/cm 2 . M-1960 was applied to clothing in the Pacific Theater to prevent 
the devastating effects of scrub typhus. The use of M-1960 as a clothing impregnant continued 
throughout the Korean and Viet Nam wars (repellent, clothing application, M-1960, 1-gal can) on a 
limited basis. Because of its tendency to irritate the skin and other reasons, M-1960 and benzyl benzoate 
were not well accepted by personnel. M-1960 also had a disagreeable odor that many soldiers and their 
commanders did not like. 11 Their effectiveness as repellents was only useful if the uniforms were 
impregnated. For this reason efforts were undertaken to replace M-1960 in 1974. A new effective 
clothing repellent was not fielded until 1991, when permethrin came into use. 

The U.S. Military recognized in the late 1980s that it needed to replace M-1960 and that the standard 
topical repellent (75% alcohol solution of deet) had drawbacks. A major research program centered at the 
Letterman Army Institute of Research in San Francisco raised new possibilities for repellent formulation 


2006 by Taylor & Francis Group, LLC 



264 


Insect Repellents: Principles , Methods, and Uses 


and generated greater expectations for personal protection products among the military entomological 
community. These expectations were translated in 1987 to a powerful document known as Joint Service 
Operational Requirement (JSOR). This document listed 11 specific requirements that ranged from 
effectiveness to compatibility with nuclear decontamination procedures. Based on preliminary results in 
the laboratory, the document set the goal as a topical repellent with 12 h of protection and a clothing 
repellent that would last at least 30 d. This system was to be operational by 1989. 12 These criteria were 
met with the adoption of a deet extended wear topical repellent (a formulation containing 33% deet in a 
polymer base) and permethrin impregnated work uniforms. Although deet is a useful clothing repellent, 13 
its volatility shortens duration on clothing and its plasticizing effect makes it inconvenient to use. The 
development of permethrin applied to clothing was a great breakthrough in the field, providing an active 
ingredient that was bound onto fibers in a way that resisted wetting, washing, and hard wear. Thanks to 
the low toxicity of permethrin and its high affinity for cloth, skin absorption from treated cloth was low 
enough to prevent anything close to toxic levels in humans wearing the clothing. 

Work on synthetic pyrethroids started in 1948. Schechter and LaForge 14 synthesized the first synthetic 
pyrethroid, allethrin, in 1949 and it was marketed in 1952. In the 1970s, synthetic pyrethroids were 
developed that were more effective and more persistent than the pyrethroids of natural origin. They 
resisted degradation when exposed to sunlight and were effective without a synergist. Scientists tested a 
number of these synthetic pyrethroids as clothing repellents including allethrin (1949), tetramethrin 
(1965), resmethrin (1967), and permethrin (1972). 15 

Permethrin is a third-generation synthetic pyrethroid. The early history of permethrin development 
involved tests on mosquitoes conducted by the U.S. Army and U.S. Air Force. Tests showed that when 
lightweight uniforms were treated until moist (approximately 3 oz) the permethrin alone (0.5%) gave 
97.7% percent protection from mosquitoes and 99.9% protection when used in combination with topical 
application of deet to exposed skin. Two detergent washings did not diminish mosquito repellent and 
killing action of permethrin-treated uniforms. When used as a repellent, permethrin is applied to outer 
clothing (i.e., not to undergarments, hats, or socks) where it dries and bonds to the cloth fiber. This water- 
based formula is nonstaining, odorless and has exceptional resistance to degradation by sunlight (UV), 
heat, and water. When placed on clothing it will last two to six weeks (even up to one year with special 
application) and will even last through weekly launderings. 16 The EPA recommended approval for 
clothing impregnation with permethrin at a concentration of 0.125 mg/cm 2 and requested additional 
information on each proposed application process. A toxicological review that included a recommen¬ 
dation to select a formulation containing permethrin and emulsifiers, was submitted in 1987. The 
recommendation also proposed environmental controls for on site impregnation methods and storage of 
impregnated uniforms. Following EPA approval of 4 impregnation methods, the Army fielded 
permethrin as the standard military clothing repellent. 17 Although not quite the same strategy as 
application of clothing repellents, the use of treated bed nets has recently become one of the major 
solutions proposed to decrease transmission of malaria parasites by Anopheles mosquitoes. Because the 
nets do not contact the skin as closely as clothing, there is a wider range of active ingredients that can be 
used. Treatment of the nets provides more protection from biting mosquitoes in several ways. First, the 
pyrethroid insecticides are repellent as well as toxic to the insect, so that a certain number of mosquitoes 
do not approach the net. Second, mosquitoes that contact the net for any length of time are either knocked 
down or killed, preventing bites through the mesh and decreasing the chance that a mosquito will enter 
the net through a hole. Finally, mosquitoes that do manage to enter the net are eventually killed as they 
contact the treated surfaces. 

A number of other synthetic pyrethroids have been used to treat nets and have displayed a great deal of 
efficacy. For instance, the World Health Organization recommends lambda-cyhalothrin, beta-cyfluthrin, 
etofenprox, and alpha-cypermethrin as well as permethrin for the treatment of bed nets. 18 Many other 
fabric fixtures (e.g., curtains, tents) have been treated with pyrethroids in attempts to kill vectors entering 
the space, but these are not, strictly speaking, repellents. The treatment of chadors, a large cloth worn as a 
combination head covering, veil, and shawl used in Islamic societies, was shown to be as effective as 
treated bed nets. 19 


© 2006 by Taylor & Francis Group, LLC 



Repellents Used in Fabric: The Experience of the U.S. Military 


265 


Repellent Application to Fabric 

A number of factors are involved in fabric impregnation. Of primary importance is the type of cloth used. 
The materials of open mesh bed nets, for example, are vastly different from that of clothing and tent 
material and differ in their absorption of repellent. Protection from arthropods was tested using several 
untreated clothing fabrics from the latter part of 1943 until early in 1945, both in the laboratory and in the 
field. Some of these fabrics afforded a great deal of protection in preventing mosquito bites because of the 
nature of their weave. Byrd cloth, a cotton twill designed during World War II to be strong, light-weight, 
and wind-proof) gave almost complete protection. The khaki cotton twill (8 oz) was only moderately 
effective, and the herringbone twill (standard 8 oz) gave little or no protection. 1 ' 20 This may have been 
due, in part, to the fabric geometry. For instance, cotton broadcloth has a higher yarn count, finer yarns, 
and a higher twist count than do cotton twill or cotton poplin. Cotton twill has lower yarn interlacings that 
provide greater yarn mobility with larger inter-fiber and inter-yarn spaces, allowing for more 
absorbency. 21 Washing of these fabrics actually caused a slight increase in resistance to insect bites. 
Also, biting was eliminated or reduced even with highly permeable fabrics when they were removed 
from direct contact with the body by coarsely woven, open mesh netting materials. 

When selecting material for impregnation, multi-filament synthetic materials (polyester, nylon) are 
preferred for a number of reasons. First, they are generally cheaper than cotton materials. The synthetics 
are also easier to impregnate with active ingredients, even though they absorb less of the solution. A 
given dose of pyrethroids is more effective on the synthetics, less insecticide is lost during washing and 
drying, and they are generally more durable. 22 Retention of repellent material in fabric is also a concern. 
Early repellents would wash from clothing materials rapidly. When solutions of dimethyl phthalate, 
indalone, Rutgers 612, or Formula 6-2-2 were applied to cotton twill cloth, protection lasted for only 
three days. Dimethyl phthalate and Rutgers 612 were ineffective after two to three risings in water. 
Indalone and dibutyl phthalate were slightly better, lasting through three to six rinsings. 21 ' 23 Current 
materials have been extensively evaluated to determine their resistance to wash and wear. Researchers 
conducted studies to determine the longevity of permethrin-treated uniforms under experimental 
conditions similar to the environmental conditions encountered in heavily forested tropical areas of 
the world. 24 ' 25 Permethrin-treated 50% cotton/50% nylon fabric provided 95% or better protection for six 
weeks against Aedes aegypti. On 100% cotton cloth, the same treatment lasted for five weeks. Although 
the concentration of permethrin remained sufficient to prevent biting for many weeks, its “knockdown” 
effect dissipated almost completely within two weeks on either the cotton/nylon blend or on 100% 
cotton. Untreated cotton fabric provided 80-92% protection while cotton/nylon provided only 80% 
protection. Wash and wear tests were also conducted to determine the effectiveness of permethrin against 
the Lone Star tick ( Amblyomma americanum). Complete protection against the Lone Star tick persisted 
through 132 h of wear and four washings when permethrin was applied at an original concentration of 
0.125 mg/cm 2 . Wearing clothing apparently degraded the permethrin treatment much more slowly than 
washing. After 132 h of wear, the treated clothing retained 95% of its effectiveness, but the treatment was 
only 49% as effective after four washings. 26 The longevity of permethrin in fabrics, based on the above 
information, is dependent on the type of material used, the duration of “weathering," and the method used 
to apply the repellent. Gupta and his colleagues 25 evaluated these parameters after 28 days of weathering 
and determined that the individual dynamic absorption (IDA) method of application was more protective 
than the aerosol spray method for both cotton and cotton/nylon blends. 

Mesh bed nets, head nets, and jackets were improved by the addition of repellents. Early repellents 
such as dimethyl phthalate and Formula 6-2-2 provided little protection when applied to mesh head nets 
of 2, 8, and 10 meshes/in. 27 However, with the introduction of repellents that were effective at a distance, 
such as deet, mesh materials were modified from 21-28 meshes/in to 4 meshes/in. 28 The wider mesh 
treated with deet greatly improved ventilation and visibility without sacrificing protection from biting 
arthropods. The amount of diluted insecticide emulsion absorbed depends on the type of material used for 


2006 by Taylor & Francis Group, LLC 



266 


Insect Repellents: Principles , Methods, and Uses 


bed nets. Relevant aspects of the textiles are weight, fabric weave, and size and surface area of net. The 
fabric-insecticide interaction depends on the type of material, as well as the formulation and dosage of 
insecticide. Cotton absorbs more emulsion, but insecticides such as permethrin, lambdacyhalothrin, and 
alphamethrin are less insecticidal on cotton than on the synthetics (an exception is deltamethrin). 
Generally, the greater amount of absorption on cotton balances the greater insecticidal effect on 
synthetics, so that a single formulation of insecticide can be used to treat either kind of cloth. 22 

Permethrin treatment of tents also provided long-lasting protection against mosquito bites. Tents 
treated with a 1% oil-based formulation of permethrin provided 82-93% protection from mosquito bites 
for the 6 weeks that the test was conducted. 29 The finishes on materials are also important. There are 
numerous dyes, binders, resins, lubricating, and softening agents that are involved in fabric finishing, any 
of which can have an effect on or be affected by repellents. Scientists have conducted tests to determine 
the effect of the repellents M-1960, benzyl benzoate, and deet used in combination with water repellents, 
detoxicants for chemical warfare agents, and fire retardants. Quarpel, a water repellent, increased the 
resistance of M-1960 and deet to rinsing. It also improved the duration of mosquito repellency of M-1960 
by 5-6 d and of deet by 4 d. It increased the tick repellency of benzyl benzoate by l-6d. The chemical 
agent detoxicant, XXCC 3 , decreased the duration of effectiveness of M-1960 to mosquitoes by 13 d and 
to ticks by 2 d. However, both M-1960 and deet increased the resistance of XXCC 3 treated cloth to 
mustard gas. The fire retardant decreased the mosquito repellency of M-1960 by 5 d. 30 In another test, an 
increase in the water repellent, amino silicone, increased the duration of mosquito repellency of synthetic 
pyrethroids including permethrin, deltamethrin, and lambdacyhalothrin. 3 'Almost any repellent can be 
used as a fabric treatment. The most efficient method is to saturate the clothing with a 5% emulsion in 
water or a 5% solution in a volatile solvent. 32 33 Some of the more effective repellents for this method are 
permethrin, deet, indalone, dimethyl carbate, dimethyl phthalate, IV-butylacetanilide, and benzyl 
benzoate. The major disadvantage of this method is that the repellency of the products diminishes 
with washing, requiring retreatment for continued effectiveness. 34 

Until recently, application of repellent material to fabric was performed primarily on an individual 
basis. With the advent of relatively safe repellents and improved technology, it is now possible to pre¬ 
treat fabric in the factory. This has allowed an increase in the commercial sales of insect repellent 
clothing. The EPA approved four methods of clothing treatment with permethrin in 1990. These included 
the individual dynamic absorption (IDA) kit, an emulsifiable concentrate designed for application with a 
two-gallon sprayer, an aerosol spray can, and factory treatment. Since July 2003, a commercial line of 
factory permethrin-treated clothing, including children’s clothing, has been EPA-registered and 
marketed to the general public. This product has a permethrin concentration of 0.52%+10%. This 
EPA-approved application rate is equivalent to 0.125 mg/cm 2 ± 10% in the ACU military nylon/cotton 
rip-stop fabric. The registration does not limit the type or weight fabric that can be impregnated and 
marketed. 35 

Hand Application 

The simplest method for individual clothing treatment is to place about a dozen drops of a liquid toxicant 
into one gloved hand, rub the gloved hands together, and then rub the material lightly on the socks and 
other clothing. More material should be applied along all openings of the clothing, such as inside the 
neckband, on the fly, and on the cuffs of trousers. 32 

Barrier Method 

With the barrier method, the liquid materials are applied only to the openings of the clothes — inside the 
neckband, on the fly, on the cuffs of the shirt and trousers, inside the waistband, on the socks above and 
inside the shoes. The material may be applied by daubing, with a sprayer, or by drawing the mouth of the 
bottle along the cloth to apply a thin layer about 1-2 cm wide. If one is not going to be crawling about on 


© 2006 by Taylor & Francis Group, LLC 



Repellents Used in Fabric: The Experience of the U.S. Military 


267 


the ground, nearly complete protection from ticks can be obtained by smearing the repellent only on the 
socks above the shoe tops and on the bottoms of the trouser legs. 1 

Spray Method 

Repellents may be applied to the clothing by the spray method for protection from mosquito bites. Special 
precautions should be taken to spray the openings to the clothes. Permethrin-based insecticide sprays 
available in the U.S. can be applied to clothing. Applications should be made outdoors, spraying each side 
of the fabric for 30-45 s or just enough to moisten it. The fabric should dry for 2-4 h before use. 1 

Dust Method 

The solid repellents may be applied to the clothing as 5% dusts, 30-60 g being required to treat the entire 
inner surface of a jacket, trousers, and socks. Only the socks and trousers need treatment if contact with 
chigger-infested ground and vegetation will be restricted to the feet and legs. Retreatment is required 
each day of exposure. Sulfur dust used in this manner will also provide protection, but should be applied 
to the skin of the legs and arms as well as to the clothing. 1 

Immersion Method 

Complete protection can be obtained by immersing field clothing in a solution or water emulsion of the 
repellent. Although permethrin is the standard, some other compounds are also used, such as benzyl 
benzoate and dibutylphthalate. 36 ' 33 Treatments with these compounds are less resistant than permethrin 
to leaching from exposure to water. About 2 g of toxicant per square foot of cloth should be used, or 60 g 
for a jacket, trousers, and socks of medium size. The underwear should not be treated. An older treatment 
called for the repellent to be dissolved in dry-cleaning fluid, calculating that 1-2 L were sufficient to soak 
a heavy cotton uniform. Although satisfactory emulsions can be made with soap, the most practical 
method is to prepare a concentrate by dissolving one of the following emulsifiers in the repellent, using 
10 parts of the emulsifier to 90 parts of the repellent: Stearate 61-C-2280 (a polyalkylene glycol stearate); 
Tween 60 (sorbitan monostearate, polyoxyalkylene derivative); Tween 80 (sorbitan monooleate, 
polyoxyalkylene derivative); polymerized glycol, monostearate, monooleate or monolaurate; Span 60 
(sorbitan monostearate) and Tween 60, equal parts; Span 80 (sorbitan monooleate) and Tween 80, equal 
parts. The final emulsion in which the clothing is dipped can be prepared by adding 250 mL of the 
concentrate to 4 L of water. It is best to agitate vigorously one part of the concentrate in two or three parts 
of water to form a creamy emulsion and then dilute with the remainder of water, using moderate 
agitation. One gallon of emulsion is sufficient to dip a set of field trousers, shirts, and socks. After 
dipping, the garments should be lightly wrung out and then allowed to completely dry before wearing. 32 

The application procedure for the individual dynamic absorption kit is specifically designed to be 
simple for soldiers in the field. First, the trousers and shirt of the field uniform are each tightly rolled, 
holding them in place with string provided in the kit. Three-quarters of a canteen cup of clear water is 
placed into a plastic bag sized to exactly hold one uniform. Ten milliliters of 40% permethrin is poured 
into the bag and it is shaken two times to mix. The bag is then unzipped and the clothing is placed in the 
bag. The bag is then closed, shaken, and allowed to sit for at least 2.5 h or more. When clothing is 
removed, it is allowed to hang for 3 h or until dry. When dry, the garments are ready to wear. One 
treatment is effective for life of the uniform. It is advised that underwear and cap not be treated. 37 

Factory Pretreatment 

Factory pretreatment of uniforms and clothing in the U.S. is limited to permethrin. There are currently 
two methods of factory treatment of uniforms and other items of clothing. End item treatment is similar to 


2006 by Taylor & Francis Group, LLC 



268 


Insect Repellents: Principles, Methods, and Uses 


the immersion method in that the repellent is impregnated after the clothing has been fabricated. 
Materials to be impregnated are placed into a washing machine containing an emulsion of permethrin. 
When the cycle is complete, the materials are placed into a dryer. When the drying operation is complete, 
the clothing is packaged and shipped. The second method is a pad application to dry cloth prior to 
fabrication often called the “wet on dry” process. 3X Repellent finishes are applied to dyed fabrics after the 
drying step. The pad application process results in a more uniform distribution of the repellent throughout 
the cloth (see Table 13.1). 62 The perceived problem with this method is that cutting of the fabric post 
application increases the exposure of garment workers to airborne levels of permethrin. 

The amount of repellent added to fabric is usually expressed as a weight percentage based on original 
fabric weight rather than as a concentration. The amount of repellent solution or emulsion applied is 
referred to as “wet pickup” and the relationship can be expressed as: 


Percent of weight “wet pickup 


weight of solution applied 
weight of dry fabric 


X 100 


(13.1) 


The actual amount of repellent applied to the fabric is referred to as “weight add-on” and can be 
expressed with the following equation: 

Percent concentration in solution (wt/wt) X Percent wet pickup 

Percent weight add-on =- (13.2) 

& 100 

Permethrin is added to uniform fabric at a target concentration of 0.125 mg/cm 2 +10%. This is 
equivalent to 0.52 (0.47-0.57) percent weight add-on. 40 


Long-Lasting Surface Treatment of Fabrics with Insecticides 

Specific polymerization of permethrin onto the fiber surface (polymer-coating method) enhances 
resistance to losses of active ingredient from weathering and laundering, preserving long-term residual 
activity against arthropod vectors. Fabrics prepared in this manner are factory-treated during production 
and are thus ready-to-use. This method has been developed only recently and suggests that the nature of 
the permethrin polymerization process on the fibers’ surface is critical to insecticidal or acaricidal 
activity as well as long-term residual activity and laundering resistance. 39 Fabrics polymer-coated with 
permethrin after the dyeing process and before tailoring yielded a theoretical permethrin concentration of 
1,300 mg/m 2 . Fabrics were dried by heating at 130°C. This new method has the following 


TABLE 13.1 

Factors Affecting Chemical Impregnation of Cloth 


Fiber type 
Yam construction 
Fabric construction 
Wettability 

Pressure of squeeze rolls 

Nature and hardness of squeeze roll coverings 

Length of immersion time 

Viscosity of solution or emulsion 

Surface tension of solution or emulsion 

Temperature of solution or emulsion 

Concentration of solution or emulsion 


Fligher wet pickup with hydrophilic fibers 
Fligher wet pickup with low twist or open end yams 
Fligher wet pickup with loose construction 
Higher wet pickup with more easily wetted fabrics 
Higher pressures lead to lower wet pickups 
Harder coverings leads to lower wet pickups 
Higher wet pickup with longer immersion time 
Higher wet pickup with higher viscosity 
Higher wet pickups with faster wetting solutions 
Viscosity and surface tension change with temperature, 
changing wet pickup 

Viscosity and surface tension change with component 
concentrations, changing wet pickups 


Source: From W. D. Schindler and P. J. Hauser (Eds.), Chemical Finishing of Textile, CRC PressAYoodhead Publishing 
Limited, Boca Raton, FL/Cambridge, England, 2004. 


© 2006 by Taylor & Francis Group, LLC 



Repellents Used in Fabric: The Experience of the U.S. Military 


269 


advantages: higher efficiency; greater permanency (up to 100 launderings); homogeneity of treatment; 
and less opportunity for generation of waste and occupational exposure. 

Olyset" Technology and Production 

This method involves the mixture of permethrin and resin, which is then extruded at high temperature as 
a fiber of appropriate thickness to include in the textile. The fiber is knitted into a net roll and the net roll 
is stitched into Olyset Net. The most important characteristic to be checked for the quality of Olyset Net 
is the dynamic behavior of permethrin in the fiber. This characteristic can be predicted by measuring the 
bleeding (release) profile of permethrin. This can be determined by repeated washing of the net with 
acetone to remove permethrin from the surface of fiber and repeated heating of the washed net to 
accelerate permethrin diffusion to the surface from inside the fiber. The repetition of washing and heating 
eventually results in a constant bleeding of permethrin, which suggests an equilibrium between bleeding 
speed from the surface of fiber and diffusion speed of permethrin within the fiber. 39 

Industrial Production of Long-Lasting Insecticide-Treated Yarns 

The technology for manufacture of long-lasting insecticide-treated yarns consists of two main elements: 
the machinery and the material. The specialized machinery includes a dosing system and feeder, which is 
designed and installed to fit the spinning equipment used by the fiber manufacturer. This system allows 
chips, granulates, powders, etc. to be added to the polymer before the fiber is extruded. The material 
itself, known as the “masterbatch,” is the concentrated active ingredients mixed with the polymer, 
usually in form of granules. In the production of fibers for mosquito nets, the insecticide is added with 
precision and adapted to the spinning conditions to keep an acceptable yield during production of the 
fiber. The active substances employed are only those recommended by WHO for treatment of mosquito 
nets. Only substances that withstand the masterbatch temperature and spinning conditions are acceptable. 
Final materials are tested using WHO specifications. In the tests so far, alpha-cypermethrin and 
deltamethrin have shown potential to meet WHO requirements. 39 

Safety of Clothing Treatments 

The EPA is responsible for the registration of pesticides. Laboratory testing for acute (short-term) and 
chronic (long-term) health effects are conducted prior to registration of a pesticide or repellent by the 
U.S. EPA. This safety testing, which is primarily conducted in animals, characterizes the effects of 
repellents on living organisms and helps identify potential effects on humans. Safe use limits are 
determined through the assessment of safety data and the judicious use of uncertainty factors 
(margins of error). When these materials are used according to the label directions, toxic effects are 
not likely to occur because the amount of pesticide that people and pets may be exposed to is low 
compared to the doses fed to laboratory animals. 41 A minimum number of tests must be 
performed 42 ' 43 to meet registration needs; however, more tests may be requested to insure that 
safety needs are met. 

The safety testing of permethrin has been conducted almost continuously since the 1970s. The Army 
Environmental Hygiene Agency initiated safety testing of permethrin in 1975, not long after its initial 
synthesis. Tests conducted from 1975 to 1977 44 evaluated dermal and oral toxicity, mutagenicity and 
teratology, skin and eye irritation, and determination of exposure. More testing was requested to insure the 
safety of permethrin when it was impregnated into cloth. Studies evaluating metabolites 45 and 
habituation, 46 as well as inhalation toxicity 47 were conducted in 1978. The migration of permethrin 
from cloth to skin was examined in 1982. 48 Skin sensitization tests were performed in 1985 49 and 


A registered trademark of Sumitomo Corp., Tokyo, Japan. 


© 2006 by Taylor & Francis Group, LLC 



270 


Insect Repellents: Principles, Methods, and Uses 


neurotoxicity was evaluated between 1985 and 1986. The issue of fabric-skin contact was addressed in 
1987. 50 The EPA requested a more advanced migration test conducted under varying environmental 
conditions, and the results reported in 1988. 51 Finally, the effect of laundering on permethrin retention in 
impregnated military fabrics was completed in 1988. 52 With this safety information, as well as reports of 
some 20 other studies, a registration application was submitted to the EPA in 1989, almost 15 years after the 
Armed Forces Pest Management Board identified permethrin as a potential clothing repellent for uniforms. 
EPA granted approval in 1990. 17 Safety studies continue even after registration, as new concerns arise. 

The Committee on Toxicology, of the National Research Council was tasked to perform an assessment 
of the health effects of permethrin-treated Army battle-dress uniforms in 1994. They concluded 
unequivocally that the risk to those wearing or manufacturing permethrin-treated uniforms was 
negligible at the intended treatment level of 0.125 mg/cm 2 . 53 Currently, the EPA is conducting a 
comprehensive review of older pesticides, including permethrin, to determine their health and 
environmental effects and make decisions about their future use. Although the toxicological charac¬ 
teristics of permethrin-treated clothing have been described in detail, the review evaluates compliance 
with the newer Food Quality Protection Act of 1996. 10 The draft of the permethrin Reregistration 
Eligibility Decision (RED) was completed in August of 20 05. 54 Permethrin neurotoxicity is primarily 
associated with effects at Na + channels of nerve axons. The prolongation of voltage-gated Na + current 
may be due to a direct toxicant effect on Na + channels, and represents a major mechanism of permethrin 
neurotoxicity. This causes multiple spike discharges within the nervous system that can lead to paralysis 
and death. Animal studies suggest that the two structural types of pyrethroids give rise generally to 
distinct patterns of systemic toxic effects. Type I pyrethroids, such as permethrin, produce a “T (tremor) 
syndrome” in animals, characterized by tremor, prostration and altered “startle” reflexes. 55 ' 56 In 
mammals, permethrin is not readily absorbed through the skin (less than 2% of the applied dose). 

Permethrin that is absorbed is rapidly changed to polar metabolites in the skin and liver, then rapidly 
excreted by the kidneys. Although occupational exposure to large amounts of permethrin has been 
associated with transient symptoms of itching, burning, or numbness, these symptoms have not been 
reported in consumers applying the products to clothing. 

Dermal exposures to military personnel are based on the clothing contact surface area of adults exposed 
to permethrin-impregnated clothing. The total exposed surface has been estimated at 0.85 m 2 , assuming 
that only the arms and legs were exposed to treated cloth. 54 Dermal exposures to garment workers are 
smaller (0.22 m 2 ) based on the more limited contact area that workers have as they contact the cloth. 

Permethrin was suspected as a contributor to Gulf War Illness, a protean disease syndrome 
experienced by some American veterans of the war to free Kuwait in 1990-1991. It was proposed 
that permethrin, deet, and pyridostigmine bromide (a carbamate given as a prophylactic against potential 
nerve agent exposure) might act synergistically to produce the illness. A number of investigators 
evaluated this hypothesis and concluded that although some synergy was seen, it was not sufficient to 
produce the illness. 57 ' 58 A number of concerns still exist about the potential for serious toxicity of 
permethrin-treated clothing. For instance, studies conducted by Dr. Steve Holladay and his graduate 
student indicated that the immunotoxicity of permethrin was exacerbated by coexposure to suberythemic 
(sub-sunburn) levels of UVB light. Although lethality was not an expected endpoint, at 15-25 pL topical 
permethrin a few mice (3-7%, respectively) developed central nervous system signs and died in about 
12 h; the rest of the mice showed no overt CNS toxicity. Simulated sunlight alone caused no such signs. 
Most Army permethrin skin absorption data were collected without considering UVB light as a 
modulating factor. It would be important to determine how low, moderate or high cis urocanic acid 
(cUCA) skin content may change systemic permethrin levels, 59 ’ 60 because both cause a marked decrease 
in the cutaneous hypersensitivity reaction. 

Another concern was presented to the Department of Pesticide Regulations in California where it was 
suggested that permethrin-impregnated garments should be registered as pesticides because residue 
discharges by treatment plants may be posing significant risks to aquatic organisms. 61 This concern was 
addressed in the recent RED for permethrin. 54 The use of permethrin-treated clothing by pregnant 
women is also a concern. 


© 2006 by Taylor & Francis Group, LLC 



Repellents Used in Fabric: The Experience of the U.S. Military 


271 


Clearly, there is an important place for clothing repellents and repellent products applied to other fabrics. 
In essence, these products protect the skin by treating its covering, just as might be done on the hair of a 
horse. Many improvements in the products available can be imagined, but permethrin impregnation of 
military uniforms has certainly been a success for the United States. In the future, new chemicals and new 
formulations may make clothing repellents safer, longer lasting, and more effective. 

References 

1. B. V. Travis and F. A. Morton, Treatment of clothing for protection against mosquitoes, Proc. 
33rd Annual Meeting N.J. Mosquito Extermination Assoc., 65, 1946, http://lrs.afpmb.org/ 
419YHlE45H5T7CY76RKQ86ZV/arc/al_06_tit_fetch/4/3662 

2. A. H. Madden, A. W. Lindquist, and E. F. Knipling, Tests of repellents against chiggers, J. Econ. 
Entomol., 37, 283, 1944. 

3. G. T. McCullough and H. A. Jones, Deposition of miticides on cloth, Report No. Ill, National 
Research Council Insect Control Committee, 1945. 

4. T. F. Whayne, Clothing, in Two Hundred Years of Military Medicine, R. C. Engelman and R. J. T. Joy 
(Eds.), Ft. Detrick, MD: U.S. Army Med. Dept., Historical Unit, 1975, Chap. 3. 

5. U.S. Department of Agriculture, Progress Report of Investigations on Insect Repellents, Orlando, FL: 
Bureau of Entomology and Plant Quarantine, 1953. 

6. E. J. Hansens and H. B. Weiss, Entomology in New Jersey, New Brunswick, NJ: Rutgers University 
Dept, of Entomology, 1954. 

7. U.S. Army, Malaria Control in the Army, War Department Technical Bulletin TB MED 164, 
Washington, DC, 1945. 

8. D. Eckroth et al. (Eds.), Kirk-Othmer Concise Encyclopedia of Chemical Technology, 4th ed., New 
York: Wiley, 1991. 

9. J. R. Reigart and J. R. Roberts, Recognition and Management of Pesticide Poisonings, 5th ed.. 
Certification and Worker Protection Branch, Field and External Affairs Division, Office of Pesticide 
Programs, Washington, DC: U.S. EPA, 1999. 

10. U.S. Environmental Protection Agency, Status of Pesticides in Registration, Reregistration, and 
Special Review (Rainbow Report), Special Review and Reregistration Division, Office of Pesticide 
Programs, Washington, DC: U.S. EPA, 1998. 

11. E. Evans, Personal Communication: Conversation on the use of M-1960 in Korea, Manager, 
Entomological Sciences Program, Directorate of Occupational Health Sciences, U.S. Army Center 
for Health Promotion and Preventive Medicine, APG, MD, 21010, 2005. 

12. U.S. Army, Memorandum: Joint Service Operational Requirement (JSOR) for Insect/Arthropod 
Repellent System, Ft. Monroe, VA: U.S. Army Training and Doctrine Command, 1987. 

13. U.S. Department of Health and Human Services, Pre- and post-travel general health recommendations, 
protection against mosquitoes and other arthropods, Travellers’ Health: Yellow Book: Health Informa¬ 
tion for International Travel, Atlanta, GA: Centers for Disease Control and Prevention, 2005, Chap. 2. 

14. M. S. Schechter. N. Green, andF. B. LaForge, Constituents of pyrethrum flowers XXIII, Cinerolone and 
synthesis of related cyclopentenolones, J. Am. Chem. Soc., 71(3), 165, 1949. 

15. G. W. Ware and D. M. Whitacre, The Pesticide Book, Willoughby, OH: MeisterPro Resources, 2004. 

16. Anonymous. Permethrin tick-killers provide superior protection, Lyme Times, 29, 10, 2000. 

17. U.S. Environmental Protection Agency, Insect/Arthropod repellent fabric treatment formulations 
containing permethrin for military use. Memo from Roger Gardner, Health Effects Division, to 
George LaRocca, Registration Division, Office of Pesticides and Toxic Substances, Washington, DC: 
U.S. Environmental Protection Agency, 1990. 

18. World Health Organization, Report of the Fourth Meeting of the Global Collaboration for Development 
of Pesticides for Public Health, Communicable Disease Control, Prevention, and Eradication, Geneva: 
WHO Pesticide Evaluation Scheme (WHOPES), 2004. 

19. M. W. Rowland et al., Permethrin treated chaddars and top sheets: Appropriate technology for 
protection against malaria in Afghanistan and other complex emergencies, Trans. Royal Soc. Trop. 
Med. Hyg., 93,465, 1999. 


© 2006 by Taylor & Francis Group, LLC 


272 


Insect Repellents: Principles, Methods, and Uses 


20. U.S. Army, Mosquito Repellency of Eight Standard Quartermaster Fabrics, Camp Ellis, IL: 363rd. 
Med. Camp Detachment (Lab), 1944. 

21. M. Raheel and E. C. Gitz, Effect of fabric geometry on resistance to pesticide penetration and 
degredation, Arch. Environ. Contamin. Toxicol., 14, 273, 1985. 

22. World Health Organization, Guidelines on the use of insecticide treated mosquito nets for the prevention 
and control of malaria in Africa, CTD/MAL/AFRO/97.4, Geneva: Roll Back Malaria Department, 
1997. 

23. D. J. Fryauff, M. A. Shoukry, and H. A. Hanafii, Contact toxicity of permethrin-impregnated military 
uniforms to Culex pipiens (Diptera: Culicidae) and Phlebotomus papatasi (Diptera: Psychodidae), 
J. Am. Mosq. Control Assoc., 12, 84, 1996. 

24. R. J. Gupta et al., Effects of weathering on fabrics treated with permethrin for protection against fabrics, 
J. Am. Mosq. Control Assoc., 5, 176, 1989. 

25. R. J. Gupta et al., Resistance to weathering in fabrics treated for protection against mosquitoes (Diptera: 
Culicidae), J. Med. Entomol., 27, 494, 1990. 

26. C. E. Schreck, G. A. Mount, and D. A. Carlson, Wear and wash persistence of permethrin used as a 
clothing treatment for personal protection against the lone star tick (Acari: Ixodidae), J. Med. Entomol., 
19, 143, 1982. 

27. K. H. Applewhite et al.. Progress in Report of the Alaska Insect Project for 1942: Repellents and 
Protective Clothing, USDA; US Army SGO Proj. 6-65-01-1; US Navy Burned Res. Proj. NMO 05012; 
Interm Rep. 0-137, pp. 56-64, 1 June 1949. 

28. R. H. Grothaus and J. F. Adams, An innovation in mosquito-borne disease protection. Mil. Med., 137, 
181, 1972. 

29. J. D. Heal, G. A. Surgeoner, and R. Lindsay, Permethrin as a tent treatment for protection against field 
populations of Aedes mosquitoes, J. Am. Mosq. Control Assoc., 11, 99, 1995. 

30. H. Markabian et al.. The compatibility of arthropod repellents with certain functional finishes of cotton 
uniform fabric, J. Econ. Entomol., 61, 464, 1968. 

31. D. D. Amalraj et al„ Insecticide impregnated cotton fabrics of different hydrophobicity against Aedes 
aegypti. Southeast Asian J. Trop. Med. Pub. Hlth., 27, 617, 1996. 

32. C. N. Smith, I. H. Gilbert, and H. K. Gouck, Use of Insect Repellents. ARS-33-26, Agricultural Research 
Service, Washington, DC: U.S. Department of Agriculture, 1960. 

33. B. R. Critchley, Insect repellents, PANS, 17, 313, 1971. 

34. R. C. Bushland, Tests against chiggers in New Guinea to develop a practical field method for 
impregnating uniforms with dimethyl phthalate for scrub typhus prevention. Am. J. Hyg. , 43,219,1946. 

35. U.S. Environmental Protection Agency, Notice of Pesticide Registration, Buzz Off Insect Shield 
Apparel, LLC, EPA Registration No. 74843-2, Office of Pesticide Programs, Washington, DC: 
Registration Division, (H7505C), 2003. 

36. C. Lanigan and M. Wheeler, Travel with Children, 4th ed., Oakland, CA: Lonely Planet Publications, 
2002 . 

37. Armed Forces Pest Management Board. Personal protective measures against insects and other 
arthropods of military significance, Technical Information Guide No. 36, 2003. 

38. W. D. Schindler and P. J. Hause, Insect Resist and Mite Protection Finishes, Chemical Finishing of 
Textiles, Boca Raton. FL: CRC Press, 2000, Chap. 16. 

39. World health Organization, Report of the Fourth Meeting of the Global Collaboration for Development 
of Pesticides for Public Health (GCDPP), World Health Organization, Communicable Disease Control, 
Prevention and Eradication, Geneva: WHO Pesticide Evaluation Scheme (WHOPES), 2004. 

40. W. D. Schindler and P. J. Hause, Chemical Finishing Processes, Chemical Finishing of Textiles, Boca 
Raton. FL: CRC Press, pp. 8-11, 2000, Chap. 2. 

41. U.S. Environmental Protection Agency, Permethrin, National Pesticide Telecommunications Network, 
Corvallis, OR: Oregon State University, 1997. 

42. U.S. Environmental Protection Agency, Regulating Pesticides: Biochemical Pesticide Test Guidelines, 
Washington, DC: Office of Pesticide Programs, 2004. 

43. U.S. Environmental Protection Agency, Regulating pesticides: Data requirements, Washington, DC: 
Office of Pesticide Programs, 2003. 


© 2006 by Taylor & Francis Group, LLC 



Repellents Used in Fabric: The Experience of the U.S. Military 


273 


44. L. W. Metker et al.. Toxicological evaluation of 3-(phenoxyphenyl) methyl ( + )-cis, trans- 3-(2,2- 
dichloroethenyl)-2,2-dimethylcyclopropane carboxylate (Permethryn), Report No. 51-031-76, 
Aberdeen Proving Grounds, MD: U.S. Army Environmental Hygiene Agency, 1977. 

45. J. A. Gere and R. E. Boldt, Determination of Urine Metabolite Levels Following Inhalation of the 
Insecticide Permethrin in Rats, Report No. 75-53-0053-79, Ft. Sam Houston, TX: U.S. Army Health 
Services Command, 1978. 

46. R. A. Sherman, Preliminary Behavioral Assessment of Habituation to the Insecticide Permethrin, 
Report No. 75-51-0026-79, Ft. Sam Houston, TX: U.S. Army Health Services Command, 1978. 

47. L. W. Metker, Subchronic inhalation toxicity of 3-(phenoxyphenyl) methyl (+ )-cis, trans- 3-(2,2- 
dichloroethenyl)-2,2-dimethylcyclopropane carboxylate (Permethryn), Report No. 75-51-0026-80, 
Aberdeen Proving Grounds, MD: U.S. Army Environmental Hygiene Agency, 1978. 

48. H. L. Snodgrass and D. C. Nelson. Dermal penetration and distribution of 14 C-labeled permethrin 
isomers. Report No. 75-51-0351-83, Aberdeenb Proving Grounds, MD: U.S. Army Environmental 
Hygiene Agency, 1983. 

49. H. L. Snodgrass, Skin sensitization of the insecticide permethrin in man and the potential for 
noninmiunological contact urticaria. Report No. 75-51-0351-86. Aberdeen Proving Grounds, MD: 
U.S. Army Environmental Hygiene Agency, 1986. 

50. H. L. Snodgrass, Fabric/skin Contact from Wearing the Army Battle Dress Uniform, Study No. 75-52- 
0687-88, Aberdeen Proving Grounds, MD: U.S. Army Environmental Hygiene Agency, 1987. 

51. H. L. Snodgrass and P. A. McGreal, Migration of Permethrin from Military Fabrics Under Varying 
Environmental Conditions, Report No. 75-52-0687-88, Aberdeen Proving Grounds, MD: U.S. Army 
Environmental Hygiene Agency, 1988. 

52. H. L. Snodgrass, The Effects of Laundering on the Permethrin Content of Impregnated Military Fabrics, 
Report No. 75-52-0687-88, Aberdeen Proving Grounds, MD: U.S. Army Environmental Hygiene 
Agency, 1988. 

53. National Research Council, Health Effects of Permethrin-Impregnated Army Battle-Dress Uniforms, 
Committee on Toxicology, National Research Council, Washington, DC: National Academy Press, 
1994. 

54. U.S. Environmental Protection Agency, Permethrin: HED Chapter of the Registration Elegibility 
Decision document, PC Code 109701, Case No. 52645-51-1, DP Barcode D319234, Health Effects 
Division, Office of Pesticide Programs. Washington, DC: U.S. Environmental Protection Agency, 2005. 

55. M. J. Clark and F. Matsumura, Two different types of inhibitory effects of pyrethroids on nerve Ca + + 
and Ca ++ /Mg ++ ATPase activity in the squid Loligo pealei. Pesticide Biochem. Physiol, 18, 180, 
1982. 

56. D. J. Ecobichon, Toxic effects of pesticides, in Toxicology: The Basic Science of Poisons, M. O. Amdur, 
J. Doull, and C. D. Klaassen (Eds.), New York: Pergamon Press, pp. 643-689, 1991. 

57. A. W. Abu-Qare and M. B. Abou-Donia, Combined exposure to DEET (7V,(V-diethyl-m-toluamide) and 
permethrin: pharmokinetics and toxicological effects, J. Toxicol. Environ. Health B, Crit. Rev., 6, 41, 

2003. 

58. W. C. McCain et al., Acute oral toxicity study of pyridostigmine bromide, permethrin, and DEET in the 
laboratory rat, Toxicol. Environ. Health., 50, 113, 1997. 

59. M. R. Prater et al.. Role of interferon g_ and transfoming growth factor b in depressed cutaneous immune 
responses caused by topical permethrin, Photodermatol Photoimmunol Photomedicine, 19, 287, 2003. 

60. M. R. Prater et al., Sunlight exposure, mimicked by cis-urocanic acid, increases both immunotoxicity 
and lethality of topical permethrin in C57B1/6N mice, hit. J. Toxicol., 22, 35, 2003. 

61. Anonymous, Permethrin residues from treated garments could lead to violations of ecotoxicity 
standards, California sanitation official warns, Pesticide.net 2:1, 2004. 

62. W. D. Schindler and P. J. Hauser (Eds.), Chapter 2: Chemical finishing process, in Chemical Finishing of 
Textile, Boca Raton, FL/Cambridge, England: CRC Press/Woodhead Publishing Limited, pp. 7-28, 

2004. 


2006 by Taylor & Francis Group, LLC 



© 2006 by Taylor & Francis Group, LLC 



14 _ 

Plant-Based Insect Repellents 


Sarah J. Moore, Annick Lenglet, and Nigel Hill 


CONTENTS 

Use of Plants Throughout the Globe.276 

Natural Fumigants.276 

Effect of Natural Fumigants on Vector-Borne Disease Incidence.277 

Repellent Chemicals Identified in Plants.278 

Alkaloids.279 

Phenols.279 

Terpenoids.280 

The Best-Known Plant-Based Repellents.283 

Citronella Group Family: Poaceae.283 

Famiaceae Family.284 

Ocimum spp.284 

Hyptis spp.284 

Mentha .284 

Thymus.285 

Daniellia oliveri (Fabaceae).285 

Tagetes spp. (Asteraceae).285 

Lippia spp. (Verbenaceae).286 

Artemisia spp.286 

Neem.286 

Garlic.287 

Lemon Eucalyptus Extract.288 

Essential Oils.288 

Commercial Plant-Based Insect Repellents.290 

Why and Where Plant-Based Repellents May Be Useful.291 

Desirable Qualities of Traditionally Used Plants.292 

Methods of Sourcing Candidate Plants.293 

Ethnobotanical Survey.293 

Ethics and Informed Consent.293 

Ethnographic Field Methods.294 

Evaluation and Toxicological Testing.295 

References.296 


275 


© 2006 by Taylor & Francis Group, LLC 




































276 


Insect Repellents: Principles , Methods, and Uses 


Use of Plants Throughout the Globe 

The use of plants against biting insects was first recorded among the ancient Greeks, 1 and plants are still 
used for this purpose by enormous numbers of people today. Most households in the developing world 
rely on personal protection measures of limited effectiveness, such as burning mosquito coils or leaves, 
despite the wide range of effective malaria-control measures available. 2 

The most common personal protection device is the mosquito coil. Each year, 29 billion mosquito 
coils are sold, 95% of them in Asia. 3 Mosquito coils were traditionally made with finely ground 
pyrethrum daisy ( Chrysanthemum cinerariaefolium) flowers mixed with coconut husks or sawdust. In 
recent years, household pesticides have used synthetic pyrethroids, based on the molecular structure of 
the pyrethrins contained in the pyrethrum daisy, more frequently than natural pyrethrins. 3 While both 
chemical groups possess rapid insecticidal and repellent action, 5 the synthetic pyrethroids are far more 
photostable than their natural counterparts. Nonetheless, 12,000 tons of natural pyrethrum are 
produced in Kenya each year to supply the household insecticide market. There is ample evidence 
that mosquito coils made from both natural pyrethrins and synthetic pyrethroids effectively repel 
mosquitoes. 7,8 

Expenditure on mosquito coils in developing countries is substantial, 9,10 and the cost of personal 
protection methods is a particularly important issue. Research has shown, for example, that in some areas 
of Thailand with high mosquito-biting densities, residents spent U.S.S12.50 to $25 per residence per year 
on personal protection, mainly mosquito coils, which represents a greater per capita expenditure than 
organized mosquito control in developed nations. 9 In India, as much as 0.63% of per capita income may 
be spent on mosquito control measures such as coils. 11 The use of store-bought preventive measures is 
generally higher among those of a higher economic status. In Malawi, for example, 67% of low- to high- 
income citizens use coils against mosquito bites, compared to only 16% of very low-income citizens, and 
11% of the wealthier people use repellents, whereas only 1% of their poorer counterparts use this 
method. 10 

Natural Fumigants 

Among poorer populations that cannot afford store-bought personal protection methods, natural 
fumigants are extensively used; less frequently, plants are hung around the home or rubbed onto the 
skin. A study from rural Guatemala found that more than 90% of households interviewed burned waste 
plant materials such as coconut husks to drive away mosquitoes. In Mexico, 69% of households used 
this technique, 13 whereas in Colombia 50% of people reported that they burned wet logs in metal pots to 
prevent mosquito nuisance, especially when fishing among the mangroves. 14 In the western Pacific, in 
Papua New Guinea, up to 90% of the population burned wood in the early evening, which was shown to 
repel 66-84% of the vector Anopheles karwari as well as nuisance culicines. 15 In the Solomon Islands, 
52% of people use fire to drive away mosquitoes. 1 ’ 

In Africa, the use of traditional fumigants is widespread. In rural Zimbabwe, 13% of people use plants 
and 15% use coils, 17 while in Malawi, 39% of people burn wood, dung, or leaves. 18 Up to 100% of 
Kenyans burned plants to repel mosquitoes, 19 and in Guinea Bissau 55% of people burned plants or hung 
them in their homes to keep the mosquitoes away. 20 The most commonly used plants in Africa include 
neem ( Azadirachta indica), Hyptis spp. (bushmint family), Ocimum spp. (basil family), Corymbia spp. 
(formerly Eucalyptus spp.) and Daniellia oliveri (locally known as churai), all of which were more than 
70% effective in field trials against Anopheles gambiae s.l. 21 

Fumigants are also commonly used to drive mosquitoes from houses throughout Asia, including 25% 
of the mobile populations interviewed in Thailand and Cambodia, 22 32% of households in rural 
Myanmar, 23 and 17% of households in southwestern China. 24 In Sri Lanka, 69% of families burned 


© 2006 by Taylor & Francis Group, LLC 



Plant-Based Insect Repellents 


277 


neem kernels and leaves ( Azadirachta indica) to repel mosquitoes, along with mosquito coils (54%), 
even though almost all houses are regularly sprayed with residual insecticide. 25 

Effect of Natural Fumigants on Vector-Borne Disease Incidence 

Two studies examined the link between traditional fumigant use and malaria incidence in The Gambia 27 
and Sri Lanka. 25 In Sri Lanka, the use of traditional fumigants against mosquito nuisance in was shown to 
be protective against malaria (relative risk=0.58). 26 In contrast, findings from The Gambia showed no 
significant difference in malaria incidence among children living in households that regularly used—or 
never used—smoldering Daniellia oliveri, although bednet use did offer some protection. 27 

The discrepancy in the results of these two studies may lie in the different mosquito and human 
behavior in the two regions. Importantly, the Sri Lankan study also showed that living less than 70 m 
away from cattle sheds was a protective factor against malaria, indicating that the Sri Lankan vectors 
have the potential to be diverted to bite cattle when repellents are used. The primary and secondary 
malaria vectors in Sri Lanka, Anopheles culicifacies and A. subpictus have low human blood indices 
(HBI) of 9.5% and 1.6%, respectively, showing a strong preference for nonhuman hosts. 28 The HBI of 
the Gambian vectors. Anopheles arabiensis, A. gambiae, and A. melas, are far greater at 52%, 28% and 
53%, respectively, when cattle are present, showing a stronger preference for human hosts. 29 Moreover, 
malaria transmission is more intense in The Gambia. The number of sporozoite-positive mosquitoes 
recovered in The Gambia is 5.6%, 30 compared to 1.06% for A. culicifacies , 28 The difference in the results 
of the two studies also could be related to the erratic use of churai in The Gambia. Although there is no 
quantitative data on frequency of use in either location, variations in the frequency with which fumigants 
are used may contribute to differences in the prevalence of malaria between villages, and between 
households in the same village. 31 

There is other evidence that malaria incidence is lower among those who use wood-burning stoves 
than those who do not. A study in Indonesia where Anopheles punctulatus is the main vector, calculated 
the entomological inoculation rate (EIR) (i.e., the number of malaria-infectious mosquito bites that an 
individual may receive in a year) as 0.015 in traditional housing, where people cook and sleep in the same 
room, and 0.058 in improved housing with a detached kitchen. 32 In the Solomon Islands, where species in 
the A. farauti complex transmit malaria, parasitaemia was 29.3% among those with an indoor kitchen and 
50.6% among those with an outdoor kitchen. 33 The reason for the reduction in malaria associated with 
the use of wood smoke may be that it drives mosquitoes from houses or prevents them from entering. 
Grooting 34 showed that traditional houses with poor ventilation and smoke-stained walls were three to 
four times less attractive to mosquitoes than more modern, ventilated houses, suggesting that smoke is an 
irritant. In Sierra Leone, in houses where wood was burned during the night, there were consistently 
higher numbers of Anopheles gambiae mosquitoes in window exit traps than in houses without wood 
smoke at night. Although mosquito feeding success was not inhibited, overall 42% more mosquitoes 
were captured in the nonsmoky room than in the smoky one. 35 Conversely, research in Sri Lanka by the 
same group that found the protective effect of natural fumigant use on malaria incidence showed that 
the use of mosquito coils and natural fumigants did not significantly reduce the number of indoor resting 
mosquitoes. 36 However, the authors note that the result may have been confounded by the fact that the 
use of traditional fumigants and mosquito coils was not continuously followed and could have changed 
midway through the study. 

The action of natural smokes is poorly understood. Mosquito coils and natural fumigants work over a 
larger area and produce smoke that may be insecticidal or irritant. Smoke production most likely has 
a long-range effect on mosquito host-seeking behavior. Unpublished data by Hill and Curtis showed that 
churai did not inhibit Anopheles gambiae feeding in the laboratory (where mosquitoes are in close 
proximity to hosts), while field tests demonstrated excellent (77%) repellency. 21 "’ 7 It is likely that smoke 
may mask human kairomones, particularly carbon dioxide. In addition, mosquitoes rely on heat and 
moisture in convection currents as a short-range cue for approach to hosts, 38 and these, too, may be 


2006 by Taylor & Francis Group, LLC 



278 


Insect Repellents: Principles , Methods, and Uses 


altered by combustion. Smoke production also lowers humidity by reducing the moisture-carrying 
capacity of the air. This makes mosquitoes susceptible to desiccation and reduces sensory input because 
mosquito chemoreceptors are more responsive in the presence of moisture. 39 Indeed, heat alone is 
repellent to mosquitoes. In a field trial in Bolivia with Mansonia titillans, volunteers sitting close to 
glowing charcoal received 31% fewer mosquito landings than those sitting close to a locally bought 
mosquito coil (positive control). However, the additi on of a local plant, Scheelea princes, further 
increased the protection to 69.2% when compared to the positive control, indicating that chemicals 
released from burning plants play an important role in repelling host-seeking mosquitoes. 24 

Attempts have been made to incorporate traditionally used plants into mosquito coils with limited 
success. Field trials in Thailand showed that the incorporation of neem ( Azadirachta indica ) leaves, 
citronella grass ( Cymbopogon nardus), lemon eucalyptus ( Corymbia citriodora), and Siam weed 
(Eupatorium odoratum) significantly reduced mosquito landings on volunteers compared to a blank 
coil, although they were significantly less effective than S-allethrin. 40 However, to date no plant-based 
coil has been produced that can compare with those based on pyrethrins or synthetic pyrethroids. 

The popularity of repellent smoke probably lies in its convenience. Since most households in the 
developing world use a wood-smoke cooking fire, the addition of plants requires minimum effort. 
Similarly, mosquito coils are the preferred anti-mosquito product used by low-income communities 9 due 
to their convenience and effectiveness. 41 In addition, the use of waste products such as coconut husks 
maximizes the usefulness of a resource, which is advantageous to low-income households, since they use 
plant repellents more frequently than households with higher incomes. 18 However, it must be noted that 
cost-effective alternatives to combustion of plant materials to drive away mosquitoes must be sought, as 
burning biomass fuels indoors causes 38,539,000 disability adjusted life years (DALYs) globally each 
year. 42 The combustion of plant materials releases many small particles, gases including carbon 
monoxide, nitrogen dioxide, formaldehyde, and carcinogens such as benzo-a-pyrene and benzene. 43 

The burning of plants also releases insecticidal and irritant chemicals. 44 It has been shown that the 
steam extracts of whole plants, including Mentha piperata and Ocimum sanctum, have knockdown 
effects. 45 It is therefore unsurprising that many plants containing repellent volatiles will repel mosquitoes 
when burned. However, the release of volatile material through thermal expulsion on hot metal plates is 
far superior to direct combustion in semi-field trials, improving repellency of Corymbia citriodora from 
51.3% to 74.5% compared to equivalent controls. 19 This is because fumes resulting from thermal 
expulsion have a richer compositional profile of the volatiles than fumes resulting from direct burning 46 
possibly because the volatiles are not fully oxidized and destroyed at the lower temperatures provided by 
thermal expulsion. 

A more modern application of the volatilization of plant actives has been proposed by the U.S. 
Department of Agriculture. In a series of olfactometer experiments, the Center for Medical, Agricultural, 
and Veterinary Entomology (CMAVE) has proposed the use of plant volatiles as spatial repellents, i.e., 
an inhibiting compound dispensed into the atmosphere of a three-dimensional environmental space, 
which inhibit the ability of mosquitoes to locate and track a target such as humans or livestock. 47 
CMAVE further proposed establishing repellent barriers by saturating a space with a spatial repellent. In 
olfactometer experiments, the team showed the excellent spatial repellency of catnip ( Nepata cataria), a 
member of the Lamiaceae. Its spatial repellency and ability to inhibit feeding were superior to deet. 48 


Repellent Chemicals Identified in Plants 

The many chemicals in plants are important in their defense against insects. These chemicals fall into 
several categories, including repellents, feeding deterrents, toxins, and growth regulators. Most can be 
grouped into five major chemical categories: (1) nitrogen compounds (primarily alkaloids), (2) 
terpenoids, (3) phenolics, (4) proteinase inhibitors, and (5) growth regulators. Although these compounds 


© 2006 by Taylor & Francis Group, LLC 



Plant-Based Insect Repellents 


279 


arose as defenses against phytophagous insects, many are also effective against mosquitoes and other 
biting Diptera. The fact that several of these compounds are repellent to hematophagous insects could be 
an evolutionary relict from a plant-feeding ancestor. It has been hypothesized that blood feeding may 
have arisen in some insect groups, including the mosquitoes, from plant-feeding ancestors in order to 
supplement nutrition. 49 Indeed, most extant species of mosquitoes (with few exceptions) and sand flies 
rely on blood to provide protein only for egg development; they gain energy from nectar. It is also 
possible that the fact that plant products commonly deter hematophagous insects is an evolutionary 
coincidence; however, it is very likely that many plant volatiles are deterrent or repellent because they 
have high vapor toxicity to insects. 50 The fumigant activity of some essential oils has been evaluated 
against a number of stored product insects. The essential oils of many Mediterranean aromatic plants 
were tested against the bean bruchid or bean weevil ( Acanthoscelides obtectus). Of those tested. Thymus 
serpyllum (creeping red thyme), T. vulgaris (common thyme). Origanum majorana (majoram), 
O. vulgare (oregano), Cinnamorum verum (cinnamon), Rosmarinus officinalis (rosemary), Ocimum 
basilicum (basil), Salvia officinalis (sage), Satureja hortensis (summer savory), Conundrum sativum 
(coriander) and Cumimum cyminum (cumin) all showed LC 50 (concentration to kill 50% of insects) of 
less than 10 mg/kg/' ’ Turmeric ( Curcuma longa) had fumigant toxicity against Rhyzopertlw 
dominica (lesser grain borer), Sitophilus oryzae (rice weevil), and Tribolium castaneum (red flour beetle). 
In work with the mosquitoes Anopheles culicifacies, A. stephensi, Culex quinquefasciatus, and Aedes 
aegypti, steam distillation extracts of Tagetes erecta (marigold) and Mentha piperata (peppermint) 
exhibited rapid knockdown activity. 45 Studies on vapor toxicity of plant volatiles to S. oryzae showed 
that terpenes from the plants, including menthone and menthol, inhibited acetyl cholinesterase activity. 54 
This is the same mode of action as organophosphate insecticides. 

Alkaloids 

Alkaloids are insecticidal at low concentrations and frequently toxic to vertebrates. They are nitrogenous 
organic molecules with varying structures. Their mode of action varies, but many affect acetylcholine 
receptors in the nervous system (e.g., nicotine), 55 or membrane sodium channels of nerves (e.g., 
veratrin). 56 Insecticidal examples include nicotine ( Nicotinia spp.), anabasine ( Anabasis aphylla ), 
veratrin ( Schoenocciulon officinale), and ryanodine ( Ryania speciosa). Physostigmine, which served as 
the model compound for the development of the carbamate insecticides, is an alkaloid isolated from the 
calabar bean ( Physostigma venenosum ). 57 Although these chemicals are not volatile, they may be used as 
repellents by burning plant material, either on a fire or in a mosquito coil, to create an insecticidal smoke 
that will repel the insects through direct toxicity. Tobacco is commonly used against biting insects across 
the globe 58 ; although this is highly inadvisable due to the carcinogenic effects of breathing fumes from 
burning tobacco. Alkaloids are found in large quantities in many members of the Berberidaceae, 
Fabaceae, Solanaceae, and Ranunculaceae families, all of which are used extensively as traditional 
insect repellents. 59,60 However, many are potent mammalian neurotoxins, and their use should 
be limited. 

Phenols 

Phenols, sometimes called phenolics, are a class of chemical compounds consisting of a hydroxyl group 
(-OH) attached to an aromatic hydrocarbon group. The simplest of the class is phenol (C 6 H 5 OH). The 
functions of phenols are diverse, contributing to cell wall structure, flower color, and defense against both 
vertebrate and invertebrate herbivores. Important phenolics in terms of insecticidal and repellent function 
are the flavonoids, which are characteristic compounds of higher plants. There are three important insect 
repellent flavonoid groups: (1) the flavones found in the Labatiae, Umbelliferae, and Compositae 
families which are quite new in evolutionary terms; (2) the isoflavonoids found mainly in the 
Leguminosae, e.g., the highly insecticidal compound rotenone (a potent mitochondrial poison) 61 


2006 by Taylor & Francis Group, LLC 



280 


Insect Repellents: Principles, Methods, and Uses 


present in Derris eliptica\ and (3) the tannins, which are found throughout the plant kingdom and exhibit 
toxicity by binding to proteins. 62 The large size of the phenols, however, means that they have little 
significance as repellents, due to their lack of volatility. They are also generally phagodeterrent. 63 

Terpenoids 

Terpenoids are among the most widespread and structurally diverse of the plant products. Terpenes are 
derived biosynthetically from units of isoprene (Figure 14.1), which has the molecular formula CsH 8 . 
The basic molecular formulas of terpenes are therefore multiples of isoprene (CsHg),,. As chains of 
isoprene units are built up, the resulting terpenes are classified sequentially by size as hemiterpenes, 
monoterpenes, sesquiterpenes, diterpenes, sesterterpenes, triterpenes, and tetraterpenes. 

There are several important groups in the triterpene category: triterpenes, steroids, saponins, sterolins, 
and cardiac glycosides. The widely publicized compound azadirachtin, derived from the neem tree 
(Azadirachta indica ) is a triterpenoid. Azadirachtin and saponins (also found in the neem tree) are insect 
growth regulators (phytoecdysones). Common triterpenes include ursolic and oleanic acid, limonins and 
cucurbitacins. Triterpenes are the constituents of many folk remedies, particularly in Asia. Plants of the 
Asteraceae family have many members that contain these compounds, and they are widely used in 
mosquito control. 59 

Monoterpenes, both cyclic and acyclic, are major components of many essential oils and are the most 
important group to consider in terms of repelling insects. One important group of monoterpenes is the 
insecticidal pyrethrins, which are harvested from the dried heads of flowers in the Chrysanthemum genus. 
These plants are still widely cultivated in Kenya, Tanzania, Ecuador, Brazil, the former Soviet Union, 
Japan, and India for use in mosquito coils and sprays. 64 The pyrethrins are a pair of natural organic 
compounds that have potent insecticidal activity. Pyrethrin I and pyrethrin II are structurally related 
esters with a cyclopropane core (Figure 14.2). They differ by the oxidation state of one carbon, and are 
viscous liquids that oxidize readily to become inactivated. Pyrethrins are neurotoxins that attack the 
nervous systems of all insects. Pyrethrum affects the central nervous systems of all types of flying and 
crawling insects, blocking sodium-gated nerve junctions so that nervous impulses fail 65 and the insect is 
knocked down and could die. In the lowest concentrations, pyrethrum affects insect behavior by 
producing a so-called “avoidance reaction” or “excito-repellency,” which results in the insect fleeing the 
source of the chemicals. 66 

The insecticides broadly act in two ways: (1) the choreoathetosis/salivation (CS) pathway, and (2) the 
tremor (T) pathway. 5 This results in many important effects: (1) the deterrence from entering a room 
where coils are burning, (2) the expulsion of mosquitoes from within, (3) interference with host finding, 
(4) bite inhibition, (5) knockdown, and (6) death. 7 When present in amounts not fatal to insects, the 
insecticides still appear to have a repellent effect. Importantly, they have low mammalian toxicity and are 
nonpersistent, being biodegradable and also easily broken down with exposure to light or oxygen. They 
are considered to be among the safest insecticides for use around food. 

The repellent protection times of several terpenes for Aedes aegypti in the laboratory are listed in 
Table 14.1. 

Citronellal or rhodinal or 3,7-dimethyloct-6-en-l-al (CioHigO) is the main component in the mixture 
of terpenoids that give citronella oil its distinctive lemon scent. It is abundant in Corymbia citriodora, the 
lemon-scented gum, as well as several repellent plants found in the Cymbopogon genus. 



FIGURE 14.1 Structural formula of isoprene. 


© 2006 by Taylor & Francis Group, LLC 


Plant-Based Insect Repellents 


281 



FIGURE 14.2 Structural formula of pyrethrin. 


Limonene is a clear, colorless liquid at room temperature with an extremely strong smell of oranges. It 
is a chiral molecule, and as is common with such forms, its biological sources produce one specific 
enantiomer. The principal industrial source, citrus fruit, contains d-limonene [(H-)-limonene], which is 
the (R)-enantiomer. Racemic limonene is known as dipentene. As the main door constituent of citrus 
(Rutaceae), d-limonene is used in food manufacturing as a flavoring, and is added to cleaning products 
like hand cleansers to give them a lemon-orange scent. However, the (R)-enantiomer is also used as a 
botanical insecticide. Limonene is found in a huge range of plants, many of which are used as repellents, 
including Thymus vulgaris (thyme). Salvia officinalis (sage). Curcuma longa (turmeric). Acorns calamus 
(sweet flag), Corymbia citriodora (lemon-scented gum), Melaleuca alternifolia (tea tree), Ocimum 
basilicum (basil), and several species of mint (Mentha spp.). 67 

Myrcene, or P-myrcene, is an olefinic monterpene. It is obtained from the essential oil of, 
among others, the plants bay ( Lauris nobilis), verbena ( Lippia citriodora), and myrcia ( Myrcia 
gale), from which it gets its name. It can also be obtained synthetically by the pyrolysis of 
pinene. Myrcene is one of the most significant chemicals used in the perfume industry; because of 
its pleasant odor, it is occasionally used directly. More importantly, it is used as an intermediate for 
the preparation of flavor and fragrance chemicals such as menthol, citral, citronellol, citronellal, 
geraniol, nerol, and linalool. However, it is also repellent to mosquitoes (Table 14.1) and is found 
in many plants used in both traditional and commercial repellent preparations, e.g., Pelargonium 
graveolens (rose geranium), Melissa officinalis (lemonbalm), Hyptis suaveolens (wild hops), Ocimum 


TABLE 14.1 


The Repellency of Essential Oils (100% Concentration) to Aedes 
Mosquitoes 


Compound 

Duration of Protection (h) 

Terpenene 

0 

Citronellal 

<1 

Limonene 

<1 

Myrcene 

<1 

a Pinene 

<1 

Citronellol 

1-2 

Eugenol 

1-2 

Linalool 

1-2 

(3 Terpeneol 

1-2 

Geraniol 

2-3 

Citral 

2-3 


Source: From USDA, Agricultural Research Sendee United States Department of 
Agriculture Handbook, Washington, DC. 1967 


2006 by Taylor & Francis Group, LLC 



282 


Insect Repellents: Principles, Methods, and Uses 


kilimandscharicum (African basil), Mentha piperita (peppermint), and Cymbopogon nardus 
(citronella). 67 

Pinene is also a monoterpene. There are two variations: ( —)-a-pinene and ( — )-(3-pinene. As the name 
suggests, both forms are important constituents of pine resin; they are also found in the resins of many 
other conifers, and more widely in other plants including sage, M. piperata (peppermint), and Corymbia 
globulus (blue gum). 67 

Citronellol, or dihydrogeraniol, is a natural acyclic monoterpenoid. Both enantiomers occur in nature. 
( + )-Citronellol, which is found in citronella oils, is the more common isomer. ( —)-Citronellol is found 
in the oils of many aromatic plants, including Pelargonium graveolens (rose geranium), Cymbopogon 
nardus (citronella), Mentha pulegium (European pennyroyal). Citrus reticulata (tangerine), and Melissa 
officinalis (lemonbalm). 67 Its characteristic sweet lemon scent lends it to many uses in the perfume 
industry, although it shows excellent repellency to mosquitoes (Table 14.1). 

Eugenol (Ci 0 H 12 O 2 ) is an allyl chain-substituted guaiacol, i.e., 2-methoxy-4-(2-propenyl) phenol. It is 
a clear to pale yellow oily liquid extracted from certain essential oils, especially clove oil and cinnamon, 
and it is slightly soluble in water and soluble in organic solvents. Eugenol has a pleasant, spicy, clove¬ 
like taste and odor useful in perfumes, flavorings, and essential oils, and it can also be used medicinally as 
a local antiseptic and anaesthetic. It is found in a range of spicy, aromatic plants including Syzygium 
aromaticum (clove), Alpinia galanga (greater galangal), A. officinarum (Chinese ginger, lesser galangal). 
Pimento dioica (allspice), Cinnamomum verum (Ceylon cinnamon), Ocimum basilicum (basil), 
O. gratissimum (shrubby basil), O. sanctum (holy basil, Tulsi), O. kilimandscharicum (African blue 
basil), Curcuma longa (turmeric), and Laurus nobilis (bay). 67 

Linalool is a terpene alcohol with many commercial applications, the majority of which are based on 
its pleasant scent (floral, with a touch of spiciness). It is found in many flowers and spice plants as well as 
in several members of the Lamiaceae family, including Ocimum basilicum (basil) and O. americanum 
(American basil). 67 

Geraniol, also called rhodinol, is a monoterpene and an alcohol. It is the primary part of oil-of-rose and 
palmarosa oil. It also occurs in small quantities in citronella and essential oils derived from Lamiaceae. 
A clear to pale yellow oil, Geraniol is insoluble in water but soluble in most common organic solvents 
with a pleasant rose-like odor. Geraniol is a constituent of many commercial insect repellents and is 
found in many repellent plants, including Thymus vulgaris (common thyme), Ocimum gratissimum 
(shrubby basil), O. basilicum (basil), Cymbopogon nardus (citronella), C. martinii (palmarosa), 
C. winterianus (Java citronella), C. citratus (lemongrass), C. flexuosus (east Indian lemongrass). 
Pelargonium graveolens (rose geranium), Corymbia citriodora (lemon eucalyptus). Zingiber officinale 
(ginger), and Mentha longifolia (mint). 67 

Citral consists of a pair of terpenoids with the molecular formula Ci 0 H 16 O. The two compounds are 
isomers based on the position of double bonds. The trans isomer is known as geranial or citral A, while 
the cis isomer is known as neral or citral B. Citral is the major constituent of the oil of lemongrass and 
several other members of the Cymbopogon genus and several citrus plants, among them C. flexuosus (east 
Indian lemongrass), C. citratus (lemongrass), C. winterianus (Java citronella). Zingiber officinale 
(ginger), Ocimum basilicum (basil), Aloysia citrodora (lemon verbena). Citrus Union (lemon), and 
Mentha rotundifolia (applemint). 67 Citral has a strong lemon odor and is commonly used in the perfume 
industry, although it is also effective as an insect repellent. 

As can be seen, repellent monoterpenes are present in many members of the Lamiaceae, Myrtaceae, 
and Poaceae. They are present in plants to deter herbivores, and some exhibit considerable toxicity to 
insects while having low mammalian toxicity. 68 Many terpenes are volatile oils, and therefore deter 
phytophagous insects by acting in the vapor form on olfactory receptors. Plants containing terpenes may 
be used as repellents without modification by rubbing fresh leaves onto the skin to release the oils; they 
may also be bruised to release the oils then hung around the home. Other uses may be as fumigants when 
the fresh leaves are burned or the oils volatilized. 19 They are also commonly added to commercial insect 
repellents that are labelled as “natural.” 


© 2006 by Taylor & Francis Group, LLC 


Plant-Based Insect Repellents 


283 


The Best-Known Plant-Based Repellents 
Citronella Group Family: Poaceae 

This genus contains several plants that are used throughout the world as insect repellents. Originating in 
India, these rapidly growing grasses with distinctive aromatic foliage are widely cultivated throughout 
the tropics. These plants contain varying amounts of several insect repellent chemicals, although 
environmental conditions cause the amount of volatile oils to differ greatly. Repellent compounds 
contained in this group include alpha pinene, camphene, camphor, geraniol, and terpenen-4-ol. The most 
abundant repellent molecules found in the group are citronellal, citronellol, and geraniol. 67 

Cymbopogon nardus or citronella is the best-known member of the group, used in many commercial 
repellent preparations. These repellents are marketed for use on children as natural repellents that are 
perceived to be safer for their use than deet. Although its ED 50 (effective dose for 50% reduction in bites) 
is similar to that of freshly applied pure deet, 69 its longevity is far inferior to that of deet at 2 h. 70 The 
short longevity of volatile plant oils is due to their high vapor pressure and consequent rapid 
evaporation. 71 

C. martinii martinii (palmarosa) is a perennial grass that is widely distributed throughout the tropics. It 
contains between 750 and 4750 ppm geraniol, 67 which gives it a sweet scent. The oil is used in traditional 
Indian mosquito repellent preparations. 72 Field tests of palmarosa against Anopheles mosquitoes in India 
showed that the pure oil provided absolute protection for 12 h. 73 However, the tests used pairs of 
volunteers, one who lay in a cot as bait while the other collected mosquitoes from him. This methodology 
may inflate the protection time of repellents. 24 ' 74 

C. citratus (lemongrass) is also traditionally used as a mosquito repellent in India. 72 Evaluation using 
an electroantennogram showed that C. citratus elicited a spike response similar to that of deet. 75 Field 
tests in Bolivia showed that 25% C. citratus in ethanol provided 77.93% and 90.67% protection for 3 h 
against Anopheles darlingi and Mansonia spp., respectively. 24 However, laboratory evaluation has 
shown far lower repellency, at only 30 min of complete protection. 76 It is rich in citral (70%), but many 
other repellent terpenes also are present, including alpha pinene, citronellal, citronellol, and geraniol. 67 

C. winterianus essential oil has been evaluated as a mixture with 5% vanillin against Aedes aegypti, 
Culex quinquefasciatus , and Anopheles dims. It compared favorably with 25% deet, giving greater than 
6 h of protection against all three mosquito species in cage experiments. 77 Another related plant is C. 
flexuosus, which contains between 875 and 2500 ppm geraniol, 67 although it does not seem to have been 
evaluated as a repellent. Used in South Africa as a mosquito repellent C. excavatus evaluated in the 
laboratory against Anopheles arabiensis gave good protection for 2 h but declined to 59.3% protection 
after 4 h, 78 which compares favorably with C. nardus. 

In Mpumalanga, South Africa, Govere et al. 78 determined through interviews with local people that 
Pelargonium reniforme (rose geranium) is considered to be effective at repelling mosquitoes. The leaves 
of this plant release a highly pungent odor. When tested, an alcohol formulation (200 mg/mL) made from 
the fresh leaves provided 63.3% and 59.3% protection after 3 and 4 h, respectively, against laboratory- 
raised Anopheles arabiensis. 

In Europe and North America, Pelargonium citrosum is being marketed as a mosquito-repelling plant 
because the leaves release a citronella-like odor. It is said that if planted, it will repel mosquitoes within a 
0.93 m 2 area. 79 However, several field experiments have shown no protective effect for volunteers sitting 
close to the plants, compared to Aedes vexans, A. triseratus, ’ A. albopictus, and Culex quinquefas- 
ciatus. ' ' The essential oil constituents were analyzed and compared to the essential oils of the 
Cymbopogon species. It was determined that Pelargonium citrosum contains trace amounts of citronellal 
and large amounts of linalool, while this ratio is reversed in C. winterianus and C. nardus oils. However, 
linalool is repellent, and the plant contains citronellol (20.82%) and geraniol (22.57%). 79 It is possible 
that applying the essential oil to the skin or evaporating it into the air would provide protection. 


2006 by Taylor & Francis Group, LLC 



284 


Insect Repellents: Principles , Methods, and Uses 


As essential oils are only repellent in the vapor phase, it is not surprising that the unbruised leaves of the 
plant provided no protection from mosquitoes. A research group in Kenya bruised the leaves of live 
potted plants before testing them and showed a significant repellent effect with several plants against 
Anopheles gambiae. 

Lamiaceae Family 

Plants from the basil family are used commonly in east and west Africa as mosquito repellents, 84 ' 85 as 
well as in India. 72 

Ocimum spp. 

The essential oils from the species of this genus contain linalool, linalol, linoleic acid, p-cymene, 
estragosl, eucalyptol, eugenol, citral, thujone, ocimene, camphor, methyl chavicol, and oleic acid, as well 
as many other terpenes, all of which are effective repellents. 67 The genus grows rapidly under a range of 
climatic conditions, although it is best adapted to a drier climate. The essential oil of O. basilicum is 
larvicidal, killing 100% of Culex quinquefasciatus at 0.12% concentration. 86 

In Tanzanian tradition, fresh Ocimum spp., called kivumbasi, are burned, and freshly cut twigs of 
O. suave and O. canum are placed in the corners of rooms to prevent mosquitoes from entering. 87 88 
The latter method was field-tested in Guinea-Bissau, west Africa, and fresh O. canum (also known as 
O. americanum) provided 63.6% protection from mosquito biting for 2 h. 20 

In Zimbabwe, Ocimum spp. leaves are rubbed on the skin as a method of repelling mosquitoes. 89 When 
the juices from the leaves of O. suave and O. canum were spread on the legs of human volunteers, the 
proportion of female Anopheles gambiae mosquitoes that were engorged with blood was reduced by 
approximately 50%. 88 A 250 mg/mL concentration of dried O. canum leaves in ethanol provided 70% 
repellency against Aedes aegypti for 2 h. 17 In Thailand, a 25% concentration of O. canum essential oil in 
ethanol was tested on three mosquito species. This formulation provided 3 h, 4 h, and 8 h of protection 
from the bites of A. aegypti, Anopheles dims, and Culex quinquefasciatus, respectively. 77 Interestingly, 
when mixed with 5% vanillin, the protection times increased greatly for each mosquito species. Vanillin 
may act as a fixative, reducing the evaporation rates of repellents. 71 

Hyptis spp. 

In the Brazilian Amazon, Hyptis sp., what is locally known as Hortela-do-campo is traditionally burned, 
and the leaves are rubbed on the skin in order to keep mosquitoes away. 58 The plant’s repellent activity is 
associated with its strong smell. In west Africa, the fresh plant is sometimes used, or else the aerial parts of 
the H. suveolens are placed on charcoal and the resulting smoke is used to repel the mosquitoes 21 (although 
thermal expulsion of the plant volatiles actually attracted mosquitoes). 19 In Tanzania, freshly picked and 
bruised sprigs of H. suaveolens, known as in hangazimu the local language, are hung in the house to try to 
prevent mosquitoes from entering, 90 also the fresh sprigs did not cause a reduction in biting when hung in 
an experimental hut (Curtis and Lines, 1986, unpublished). In comparison, when tested in Guinea-Bissau, 
the fresh plant was able to provide approximately 70% protection from biting for 2 h. 20 The smoldering 
plant provides the most effective protection. Nicholson and Lines (1987, unpublished) showed that there 
was a ten-fold reduction in biting in the presence of hangazimu smoke. Similarly, Palsson and Jaenson 20 
showed that smoldering H. suaveolens provided approximately 84% protection for 2 h against Anopheles 
gambiae. In contrast, Seyoum et al. 19 found only a 20.8% reduction in biting. 

Mentha 

There are few published accounts of Mentha plants being used as personal protection against mosquitoes. 
In the Brazilian Amazon, the leaves are either rubbed on the skin or burned to produce smoke. 58 


© 2006 by Taylor & Francis Group, LLC 



Plant-Based bisect Repellents 


285 


Barnard (1999) 91 tested several concentrations of the M. piperita essential oil against Aedes aegypti and 
determined that at 25% and 100% concentrations the protection times were 30 and 45 min, respectively. 
Field tests of M. piperata in India against Anopheles culicifaces. An. annularis, and An. subpictus 
provided 85% protection over 11 h. 2 However, these results are inflated, as insect collectors not wearing 
repellent collected the mosquitoes from bait-individuals wearing repellent. Evaporating the essential oil 
of M. piperita at room temperature caused knockdown of several mosquito species. 45 Peppermint is 
widely grown throughout the tropics for its essential oil. The leaves yield up to 80% menthol, which 
is the insecticidal ingredient contained in this species. The plant also contains mosquito-repellent 
chemicals including menthone, limonene, beta pinene, alpha pinene, and linalool. 54 

Thymus 

Thyme oil at 100% is repellent against Aedes quadrimaculatus, Aedes aegypti , and Anopheles albimanus 
for at least 30 min when applied to cloth. 70 Most recently, varying concentrations of the essential oil of 
red thyme were tested in the laboratory against Aedes aegypti and An. albimanus. 91 At 100% 
concentration, it provided 135- and 105-min protection against Aedes aegypti and An. albimanus, 
respectively. At 25% concentration, the protection time was 45 min for both species of mosquitoes. It 
also was determined that mixtures of essential oils were, in fact, no more effective than the essential oils 
alone. A local method of protection against biting insects in the Soviet Union involved tying thyme stick. 
Thymus serpyllum, with thick cotton, drying this, and then burning it. Rubtzov tested this method and 
reported 85-90% protection for 60-90 min in the open air. 90 

Daniellia oliveri (Fabaceae) 

The local names churai, santang, and santao refer to resins and wood commonly burned indoors in 

• • • • no oc '^'7 , 

western Africa to prevent mosquitoes from entering at night. - ’ ’ In several held trials, it was 
determined to be an effective, accepted, and cheap form of personal protection. In Guinea Bissau, smoke 
from the burning bark of Daniellia oliveri reduced biting from mosquitoes by 74.7% and 77.9% in 
comparison to the control in two separate held experiments. 20 In Banjul, Gambia, santango reduced 
biting on human subjects by 77%, which was more effective than a permethrin mosquito coil but less 
effective than deet soap. 37 

Tagetes spp. (Asteraceae) 

Tagetes species contain monoterpenoid esters, 93 and their larvicidal and insecticidal activity is well 
established. 94-97 The essential oil was determined not to be a mosquito repellent. 94 However, studies in 
Zimbabwean communities showed that people use fresh plant material of T. minuta as a form of personal 
protection 17 by crushing the plant material and applying it to the skin, burning it, or simply exposing the 
whole plant. Okoth 98 tested the effectiveness of whole plant material of T. minuta against mosquitoes in 
Uganda. The held site had large numbers of Mansonia uniformis and Anopheles marshalli. Human¬ 
landing catches were performed in a tent in which 4 kg of fresh T. minuta whole-plant material had been 
placed one hour earlier, and in a control tent with no plants. Fewer mosquitoes were recorded biting and 
resting in the tent where the plant material had been placed than in the untreated tent. Preliminary 
laboratory tests also showed that the plants had repellency in a choice test, and signihcant toxicity when 
mosquitoes and plant parts were put in containers together. 

More recently, Tyagi et al." carried out cage tests using the essential oil of T. minuta. After 6 h, 86.4% 
protection was provided against Anopheles stephensi, 84.2% against Culex quinquefasciatus, and 75% 
against Aedes aegypti. Steam distillate of T. minuta evaporated at room temperature caused rapid 
knockdown of mosquitoes, including An. culicifacies and An. stephensi , 45 These results suggest that this 
plant has excellent potential as a mosquito repellent, although further testing is required. 


2006 by Taylor & Francis Group, LLC 



286 


Insect Repellents: Principles , Methods, and Uses 


Lippia spp. (Verbenaceae) 

In The Gambia, Lippia cheraliera leaves are traditionally used as mosquito repellents. L. javanica is 
commonly found in southern Africa, where it is frequently used as a repellent. 17 The leaves have a strong 
lemon smell, 100 which probably accounts for the local belief in its healing abilities. L. cheraliera is also 
burned in The Gambia as a mosquito-repellent smoke. 101 A thorough study carried out in Zimbabwe 
revealed that 29% of the population used plants, mainly by burning the leaves of L. javanica, to protect 
themselves from mosquitoes. 17 The main constituents of the essential oils of this plant are mono¬ 
terpenoids, such as myrcene, caryophyllene. linalool, p-cymenc, and ipsdienone. An alcohol extract of 
dried L. javanica leaves was tested on human subjects against Anopheles arabiensis mosquitoes in the 
laboratory. 78 The protection was 76.7% after 4 h and 59.3% after 5 h. Alcohol extracts of the dried leaves 
applied to the skin were also shown to provide 100% protection for 2 h against Aedes aegypti , 89 Work 
using the related L. uckambensis has shown that the release of volatiles from the leaves through thermal 
expulsion reduces An. gambiae biting by 49.5%. 19 

Artemisia spp. 

Members of this genus are found all over the world, from tropical India to Siberia. They are low-growing 
perennial herbs in the family Asteraceae. The plants are aromatic, can tolerate poor conditions, and 
provide good cattle fodder, and they have been used against insects for centuries. In China, bundles of 
dried Artemisia vulgaris are burned to repel biting insects. This observation led to an investigation by 
Hwang et al. 44 which revealed that A. vulgaris contains insect repellents that can be released from the 
plant by combustion. The compounds that were isolated and found to repel Aedes aegypti were camphor, 
linalool, terpenen-4-ol, a and P thujone, P pinene, myrcene, limonene, and cineol. To repel mosquitoes, 
the plants are also burned in central Asia, Bolivia, and India, ' ’ and by many Native American 
peoples, including the Shuswap, Thompson, and Blackfoot tribes. 102 Extracts of A. vulgaris are also 
highly toxic to mosquito larvae. 103 

A. absinthium (absinthe) is a native of Europe, central Asia, and Africa, yet it was only used as an 
insecticide in Europe 104 and India. 105 It is insecticidal, 106 and contains many repellent chemicals, 
including thujone, terpinen-4-ol, linalool, nerol. geraniol, -pinene, and 1,8-cineole. 107 Although it is 
reported as a mosquito repellent, 108 A. absinthium does not appear to have been evaluated 
against mosquitoes. 

Neem 

The neem tree (Azadirachta indica ) has become a focus of attention with regard to the control of 
agricultural pests, and more recently against medically important insects. It originates in India, where it 
has been used to control and repel insects for thousands of years. It has now been introduced to drier parts 
of Central and South America, Africa, Australia, and southeast Asia, notably southern China, where 
extensive plantations may now be found. Neem is widely used in its raw form as an agricultural 
pesticide, 109 and its leaves are traditionally burned to repel mosquitoes in Africa 20 and South America, 58 
while the leaves and husks are burned for this purpose in Sri Lanka. 25 The trees can grow in depleted and 
saline soil, making them an excellent method of regenerating desertified or marginal land. They are fast 
growing and can be used for a multitude of purposes besides insect control, including firewood, fodder 
for livestock, and shade. 

Extensive research has been carried out on the effect of botanical derivatives of the neem tree and its 
relatives. 110-112 Artemisia indica contains at least 35 biologically active principles, of which azadirachtin 
is the predominant active ingredient. It is found in the seed, leaves, and bark. The azadirachtin content of 
neem oil is positively correlated with its effect on insects, 113 which may be grouped into six categories: 
(1) antifeedency, (2) growth regulation, (3) fecundity suppression, (4) sterilization, (5) oviposition 
repellency or attraction, and (6) changes in biological fitness. 


© 2006 by Taylor & Francis Group, LLC 



Plant-Based Insect Repellents 


287 


The repellency of neem oil to hematophagous insects has been tested, although the results have been 
variable. Burning and thermal expulsion of the leaves produces only a modest reduction (greater than 
25%) in biting. 19 However, experiments using neem oil derived from the seeds have shown better 
protection. A field test in India with Anopheles culicifacies using 2% neem oil in coconut oil provided 
100% protection for 12 h. 114 Another field test of 2% neem oil in coconut oil provided 98.0% protection 
against An. culicifacies. 115 Caution should be exercised when interpreting the results of these two studies, 
as numbers of mosquitoes were extremely low (7.7 and 8.06 mosquitoes/man-hour, respectively, 
captured from controls). Bait subjects lay in cots, which may have given them some protection from 
host-seeking mosquitoes. In field tests with An. dims, 66.7% protection was recorded after 9 h using 2% 
neem oil diluted in mustard oil, again during trials with low landing rates of only 5.25 mosquitoes/man¬ 
hour on controls. 116 

In contrast, when Pandian and Devi 117 tested neem oil in coconut oil against Armigeres subalbatus, they 
found that it provided only 124 min of protection. In comparison, in the Bolivian Amazon with high 
densities of An. darlingi (mean 71 mosquitoes/man-hour) 2% neem oil in ethanol provided only 56.7% 
protection 3 and 4 h after application. 118 Another field test against An. darlingi was performed in the 
Venezuelan Amazon using a commercial preparation based on neem oil and citronella in a carbomer base. 
This preparation—Neemos* gel—offered 98.2% protection against An. darlingi for 8h. 119 Although 
numbers of mosquitoes were high (217/man-hour) in hand catches before the tests, the number of 
mosquitoes captured from the control was far lower, at 13.78/man-hour. This discrepancy may have been 
because each volunteer lay in a cot, and raising the feet from the ground significantly reduces the number of 
attempted feeds that an individual receives. 37 Interestingly, a field trial was conducted in India, where 
volunteers applied 2% neem oil in the same way as in the aforementioned trials, but then sat on the ground. 
The protection provided by the neem oil was only 73% in the first hour after application. 120 

The most effective result was obtained by vaporizing neem oil from mats: 5% neem oil was more 
effective at reducing both biting and numbers of resting mosquitoes than 4% allethrin on mats. 121 It has 
also been proposed that neem may be used to repel mosquitoes by adding it to kerosene for use in the 
kerosene lamps used to light homes throughout the developing world. 122 Adding 1% neem oil to 
kerosene provided up to 84.6% protection from bites. Unfortunately, the paper is not clear as to whether 
treatments and control collections were carried out on the same day, and neither is there any mention 
made of baseline mosquito numbers. When 1% transfluthrin was added to kerosene, only a 43.8% 
reduction in biting was witnessed. 123 If neem oil in kerosene is effective at repelling mosquitoes, this has 
important implications for malaria control due to the ease of application of this method. Neem oil is 
cheap and freely available throughout India and many other regions of the world. Perhaps a better way of 
releasing the volatile repellent might be to place the repellent and oil mixture above the flame and not in 
the kerosene itself. Transfluthrin (0.5%) volatized in this way provided more than 90% protection 123 and 
had the advantage that the optimal temperature for release of the repellent could be better regulated. This 
method offers considerable promise because of its extreme simplicity and convenience and the wide use 
of kerosene lamps. 

Garlic 

It is still a common misconception that eating garlic. Allium sativum, will make the skin unpalatable to 
mosquitoes, 124 a view that has been held since ancient times. 1 Garlic is still used as a repellent in 
South America (where it is hung around the neck) and in China (where it is eaten). 24 ' 58 Stjernberg and 
Berglund 125 claimed that the consumption of 1200 mg garlic per day provided significant protection 
against tick bites. However, the accuracy of this study has been contested, since the findings were 
exaggerated statistically due to the incorrect use of the collected data. 126 Conclusive evidence 
that consumption of garlic does not repel mosquitoes has been found using a double-blind 


A registered trademark of Medivet, India. 


© 2006 by Taylor & Francis Group, LLC 



288 


Insect Repellents: Principles, Methods, and Uses 


randomized trial. Garlic does in fact have insecticidal properties, “ as it contains allicin.as well as the 
repellent compounds geraniol, linalool, caffeic acid, and ferulic acid. Commercial insecticide and 
repellent preparations based on garlic are available, and are certified for use against mites, nematodes and 
mosquito larvae (Garlic Barrier AG, EPA #66352-2, from Allium Associates). 

Lemon Eucalyptus Extract 

The lemon eucalyptus extract comes from the plant Corymbia citriodora, which originates in China. 
Synonyms include Eucalyptus citriodora and E. maculata var citriodora. Chemical analysis of C. 
citriodora showed that it contained citronella, citronellol, geraniol, isopulegol, delta pinene, and 
sesquiterpenes. 129 The essential oil extract was determined to have mosquito-repelling properties 
against Aedes aegypti. Although its protection was limited to 1 h, 129 it is slightly superior to the 

. TO *7A 1 'tO 1 o 1 

protection period of essential oils of several other species of eucalyptus.” ’ ’ However, 

p-menthane-3,8-diol (PMD) was discovered as a by-product. It is a white, waxy material, semi-solid 
at room temperature, produced as a distillate after acid modification of the lemon eucalyptus oil. This 
material was determined to be highly repellent, and was given the Chinese name Quwenling, which 
means “effective repeller of mosquitoes.” 

PMD has undergone several trials in different parts of the world. Laboratory studies by Trigg and 
Hill 132 showed that 30% PMD was almost as effective as deet, the most widely available synthetic 
repellent, against Anopheles gambiae, which is the primary malaria vector in sub-Saharan Africa. It was 
determined that PMD impregnated towelettes (0.575 g) applied to the arms of human volunteers 
provided 90-100% protection against mosquitoes from laboratory-raised colonies of An. arabiensis . 133 

Field studies in China showed that the protection time from Aedes vexans and A. albopictus was 2 and 
5.5 h, respectively, when PMD was used in a 20-30% glycerol and/or alcohol formulation. 129 In 
Tanzania, 50% PMD in isopropanol provided over 6 h of protection from the local malaria vectors An. 
gambiae and An. funestusP 4 In the Bolivian Amazon, 30% PMD in an alcohol base provided 96.9% 
protection from all mosquito species for up to 4 h after application, compared to 84.8% protection from 
15% deet. 1 IX It is worth noting that 81.3% of the mosquitoes caught in the study area were An. darlingi, 
the principal malaria vector in the whole Amazon region. 

Although it is sometimes derived through synthetic means, PMD is a now a well-established natural 
product with proven field efficacy. In addition, acute toxicity studies show limited toxicity, with oral 
LD 50 (lethal dose for 50%) of 2408 mg/kg and dermal LD 50 greater than 2000 mg/kg in rats. 134 

For these reasons, the potential for commercial exploitation is high. Currently, Quwenling is available 
commercially in the U.S. and several countries in Europe. It is the only plant-derived insect repellent 
that is approved for use in disease prevention by the United States Centers for Disease Control and 
Prevention (CDC). 135 

Cymbopogon citriodora also shows promise for adaptation into low-technology applications, by 
thermally expulsing volatiles from the fresh leaves. Heating the leaves on a metal plate over a traditional 
cooking fire in western Kenya reduced Anopheles gambiae landings on occupants of a house by 74.5%, 
which is comparable to insecticidal mosquito coils. 19 


Essential Oils 

Essential oils are derived by steam distillation from plants in several families. The Lamiaceae family 
contains several well-known repellent plants, including basil ( Ocimum basilicum), mint ( Mentha spp.), 
hyptis (Hyptis suaveolens), lavender ( Lavandula spp.), sage {Salvia spp.), and thyme {Thymus spp.). The 
Myrtaceae family includes eucalyptus {Corymbia spp.) and tea tree {Melaleuca spp.), and the Poaceae 
includes citronella, lemongrass, and palmarosa {Cymbopogon spp.). Table 14.2 shows the average 
protection times of a range of essential oils against Aedes aegypti mosquitoes. 76 


© 2006 by Taylor & Francis Group, LLC 




Plant-Based Insect Repellents 


289 


TABLE 14.2 


Repellent Activity of Essential Oils (Undiluted or as 10% and 50% Dilutions) Against Aedes aegypti Mosquitoes 


Scientific Name 

Family 

English Name 

Duration (min) of Complete Repellency 

10% 50% Pure Control 

Ageratum conyzoides — 

Asteraceae 

— 

20 

30 

60 

0 

Allium sativum —L 

Alliaceae 

Garlic 

10 

40 

70 

0 

A. tuberosum —Roxb 

Alliaceae 

Oriental garlic 

0 

0 

0 

0 

Apium graveolens Linne 

Umbelliferae 

— 

0 

60 

40 

0 

Boesenbergia pandurata 

Zingiberaceae 

— 

0 

0 

30 

0 

Roxb Schltr 

Canagium odoratum —Baill 

Annonaceae 

Ylang Ylang 

0 

0 

30 

0 

Ex King 

Cedrus deodar a 

Pinaceae 

Cedar 

0 

0 

0 

0 

Citrus hystrix —Dc 

Rutaceae 

Leech lime 

10 

30 

60 

0 

C. reticulata —Blanco 

Rutaceae 

Tangerine 

0 

0 

0 

0 

Cupressus funebris 

Cupressaceae 

— 

0 

0 

10 

0 

Curcuma longa —L 

Zingiberaceae 

Turmeric 

0 

10 

10 

0 

Cymbopogon citratus —(Dc.) 

Poaceae 

Lemon grass 

0 

30 

30 

0 

Stapf 

C. nardus —(L). Rendle 

Poaceae 

Citronella 

0 

60 

120 

0 

Corymbia globulus —Labill 

Myrtaceae 

Eucalyptus 

0 

0 

30 

0 

Lavandula angustifolia 

Lamiaceae 

Lavender 

0 

0 

10 

0 

Litsea cubeba 

Lauraceae 

Litsea 

0 

0 

0 

0 

Mentha arvensis —L 

Lamiaceae 

Japanese mint 

0 

30 

50 

0 

Mentha piperita —L 

Lamiaceae 

Peppermint 

0 

0 

50 

0 

M. spicata —L 

Lamiaceae 

Spearmint 

10 

30 

30 

0 

Myristica fragrans —Houtt 

Myristicaceae 

Nutmeg 

0 

0 

30 

0 

Ocimum basilicum —L 

Lamiaceae 

Sweet basil 

0 

0 

70 

0 

O. sanctum —L 

Lamiaceae 

Holy basil 

0 

10 

60 

0 

Pelargonium graveolens 

Geraniaceae 

Geranium 

10 

40 

50 

0 

Pimpinella anisum 

Umbelliferae 


0 

0 

0 

0 

Pinus sylvestris —L 

Pinaceae 

— 

0 

40 

60 

0 

Piper betle —L 

Piperaceae 

Betel pepper 

0 

70 

80 

0 

P. nigrum —L 

Piperaceae 

Black pepper 

0 

0 

90 

0 

Pogostemon cablin —Blanco 

Lamiaceae 

Patchouli 

0 

60 

120 

0 

Sesamum indicum —L 

Pedaliaceae 

Sesame 

0 

0 

0 

0 

Spilanthes acmella —(L) Murr 

Asteraceae 

Paracress 

30 

0 

30 

0 

Syzygium aromaticum —(L) Men* 

Myrtaceae 

Clove 

30 

70 

120 

0 

Vetiveria zizanioides —Nash 

Poaceae 

Vetiver 

0 

10 

60 

0 

Vitex negundo — L 

Labiatae 

Indian privet 

0 

0 

10 

0 

Zanthoxylum limonella —Alston 

Rutaceae 

— 

30 

80 

120 

0 

Zingiber officinale —Roscoe 

Zingiberaceae 

Ginger 

0 

0 

60 

0 

Z. purpureum —Roscoe 

Zingiberaceae 

— 

0 

0 

40 

0 


Source: Adapted from Y. Trongtokit, et al., Phytother. Res., 19, 303, 2005. 


Before the discovery of effective synthetic repellents, aromatic oils were used as repellents by the 
military. Members of the British Indian army were issued a cream containing citronella, camphor, and 
paraffin, although this was only effective for 2 h. 136 

There are a few indications that essential oils prevent malaria. In 1945, Philip et al. 137 reported lower 
spleen indices in women than men in southern Madras. The authors observed that the local malaria vector 
Anopheles fluviatilis was biting men preferentially. The women of the region smeared themselves with 
turmeric (Curcuma longa), galangal (Kaempferia galanga ), and mustard oil (Brassica juncea ) before 
bathing. Tawatsin et al. 77 found that the steam distillate of turmeric plants provided 8 h of protection 


2006 by Taylor & Francis Group, LLC 



290 


Insect Repellents: Principles, Methods, and Uses 


TABLE 14.3 


Performance of Several Natural-Based Commercial Repellents Compared to 7% Deet (Skinsations) 


Product 

Ingredient(s) and Formulations 

Mean Protection 
Time (MPT) in 
Hours" 

Repellency Index b 

Neem Aura 

Aloe vera, extract of barberry, camomile, goldenseal, 
myrrh, neem, and thyme; oil of anise, cedarwood, 
citronella, coconut, lavender, lemongrass, neem, 
orange, rhodiumwood, NeemAura Naturals, Inc., 
Alachua, FL 

1.5 

0.3 

GonE! 

Aloe vera, camphor, menthol, oils of eucalyptus, 
lavender, rosemary, sage, and soybean, Aubrey 
Organics, Tampa, FL 

0.9 

0.2 

SunSwat 

Oils of bay, cedarwood, citronella, goldenseal, juniper, 
lavender, lemon peel, patchouli, pennyroyal, tansy,tea 
tree, and vetivert, Kiss MyFace Corp., Gardiner, NY 

1.5 

0.3 

Natrapel 

Citronella (10%), Tender Corp., Littleton, NH 

2.3 

0.5 

Bygone 

Oils of canola, eucalyptus, peppermint, rosemary, and 
sweet birch, Lakon Herbals, Inc., Montpelier, VT 

1.5 

0.3 

Bite Blocker 

Glycerin, lecithin, vanillin, oils of coconut, geranium, 
and soybean (2%), Consep, Inc., Bend, OR 

7.2 

1.5 

Skinsations 

Deet (A0V-diethyl-3-methylbenzamide, 7%), Spectrum 
Corp., St. Louis, MO Spray 

4.8 

1.0 

Repel 

lemon eucalyptus insect repellent lotion. Oil of lemon 
eucalyptus (65% p-menthane-3,8-diol [PMD]) (26%), 
Wisconsin Pharmacol Comp., Inc., Jackson, WI 

7.6 

1.7 

MosquitoSafe 

Geraniol 25%, mineral oil 74%, Aloe vera 1%, Naturale, 
Ltd., Great Neck, NY 

3.1 

0.6 


a Three mosquito species are Aedes (Stegomyia) albopictus, Culex nigripalpus, and Aedes (Ochlerotatus) triseratus. 
b Derived by dividing “Mean” by “Mean” for Skinsations (4.8 h). 

Source: Adapted from D. R. Barnard and R. D. Xue, J. Med. Entomol., 41, 726, 2004. 


against An. dims, and a hexane fraction of galangal provided 3 h of protection from Aedes aegypti in cage 
experiments. Mustard oil provided 2.1 h of protection in field tests against An. culicifacies. It is 
possible that the lower spleen indices in these women was due to their use of plant oils, especially since 
An. fluviatilis bites for only a few hours early in the evening. 140 However, it is unlikely that the burden of 
malaria in the region is reduced by the use of these oils because the mosquitoes were presumably diverted 
to biting the women’s unfortunate husbands. 

Commercial Plant-Based Insect Repellents 

Commercial botanical repellents are widely available and are based mainly on citronella, although 
several are available that use essential oils (Table 14.3). These essential-oil based repellents generally 
perform significantly less well than deet, with an average repellent protection time of between 5 min and 
2h, 141-143 a level that is not recommended for use in disease-transmission areas. 144 However, in 
scenarios where vector-borne pathogen risk is low, the short protection time of natural repellents may be 
overcome by their frequent re-application. 

Of the commercial varieties, Bite Blocker performed well, with a mean protection time of 7.2 h under 
laboratory conditions. 141 A field test showed that Bite Blocker was repellent for 3.5 h under intensive 
biting pressures from Aedes stimulans, A. canadensis, A. euedes, and A. fitchii. 145 However, it is 


Bite Blocker is a registered trademark of HOMS LLC, Clayton, NC. 


© 2006 by Taylor & Francis Group, LLC 






Plant-Based Insect Repellents 


291 


considered a third-line repellent by Health Canada, as no independent field research has been performed 
on this compound . 146 Nonetheless, it is one of the more effective plant-based commercial repellents 
available today. 

Similarly, repellents made with citronella protect for 2 h. In field tests against Aedes spp. in Canada, 
Buzz Away (5% citronella oil, plus cedarwood, eucalyptus, lemon grass, and peppermint essential oils) 
and Natrapel * 1 (citronella oil 10.0%), provided 92.5% and 65.6% protection, respectively, after 30 min. 
This level of protection fell to 64.3% and 32.4%, respectively, 3 h after application . 147 

Why and Where Plant-Based Repellents May Be Useful 

Following the development of efficient synthetic products, little attention has been paid to researching 
plant-based insecticides and repellents or to testing whether traditional preparations are effective . 90 
Funding for malaria research, especially research focusing on plants, has declined recently, and 
commercial interest is low, with biopesticides comprising only 1 % of the world pesticide market . 148 
Low commercial acceptance of biopesticides is due to several factors, including limited shelf life and 
slower action in comparison to synthetic compounds . 149 This may be a contributing factor to the lack of 
novel compounds being developed in recent decades. 

However, a review by Sukumar et al . 130 highlighted the potential of many plants in mosquito control 
through their use either as skin repellents, space repellents, insecticides, or larvicides. Along with the 
possibility of new methods of personal protection for individuals, research into botanicals may provide 
chemical skeletons for new compounds. The best-known example for this is pyrethrum, obtained from 
Chrysanthemum cinerariaefolium. More recently, several new insect repellents have been developed 
based on a piperidine skeleton, which is present in piperine, the main active chemical agent in pepper 
(Piper spp.). Piperidine is an organic compound with the molecular formula C 5 H 1 |N. It is a cyclic amine 
with a six-membered ring containing five carbon atoms and one nitrogen atom. It is a clear liquid with a 
pepper-like odor (Figure 14.3). 

During the 1970s, around 600 synthetic compounds related to piperidines were developed by scientists 
at the USDA research centers in Beltsville, Maryland, and Gainesville, Florida. The data from these 
experiments are now being re-examined using new, recently developed methodologies coupled with 
rapid screening bioassays. This interest in finding deet alternatives has been motivated by the controversy 
around the safety of deet, its low user acceptability, and its plasticizing effect. The repellent 1-piperidine 
carboxylic acid, 2(2-hydroxyethyl)-, 1-methylpropylester was developed by Bayer in the 1980s using 
molecular modelling . 150 More recently, optically active (IS, 2S)-2-methylpiperidinyl-3-cyclohexen-l- 
carboxamide (SS220) has been developed as a highly effective synthetic arthropod repellent . 151 

Traditional plant-based mosquito control products have several additional advantages. First, they are 
inexpensive and easily available, particularly if people grow them themselves. Plant-based repellents can 
be produced locally, which reduces their cost and could help boost the local economy. In the developing 
world, 80% of people are thought to rely on herbal remedies for primary health care needs . 152 Plant- 
based repellents may also be more culturally acceptable in communities with a tradition of plant use, 
where synthetic products may be perceived as unhealthy or unpleasant smelling. 

An excellent example of plants making repellent preparations more culturally acceptable was shown 
during research performed in Myanmar. Karen women on the Thai-Myanmar border use thanaka , a 
cosmetic preparation made from the pulp of the wood apple tree, Limonia acidissima. This preparation is 
slightly repellent at high concentrations and enhances the repellency of deet when the two are mixed 
together . 153 In a follow-up clinical trial, pregnant women using a mixture of thanaka and deet 
experienced a 28% reduction in incidence of falciparum malaria compared to women using thanaka 
alone (;? = 897), although this did not reach statistical significance with a log rank test . 154 The authors 


Buzz Away is a registered trade mark of Quantum, Inc., Santa Monica, CA. 

1 Natrapel is a registered trademark of Tender Corp., Littleton, NH. 


© 2006 by Taylor & Francis Group, LLC 



292 


Insect Repellents: Principles , Methods, and Uses 



FIGURE 14.3 Structural formula of piperidine. 


suggested that the combination of thanaka and deet could be useful in areas of low malaria transmission. 
Local vector mosquitoes bite early in the evening, and the prevalence of multi-drug resistant Plasmodium 
falciparum is extremely high. Thus, the use of repellents for pregnant women is strongly recommended. 
In addition, significantly more women expressed a preference for the thanaka and repellent mixture, 
compared to the repellents alone. 

Desirable Qualities of Traditionally Used Plants 

Plants developed for biocides need to be sustainable. Ideally, they will be fast-growing and naturally 
abundant or easy to cultivate. The source of the repellent should be obtained preferably from replaceable 
parts of the plant, such as the leaves or seeds, rather than parts such as roots or shoots that damage or kill 
the plant when removed. Abundance and survival after parts have been harvested is important for 
sustainability, because useful plants may become scarce due to over-harvesting if they are insufficiently 
common or robust. 155 The parts used must be available when needed or be easy to harvest and store. 

To ensure compliance, plant-based repellents need to be easy to use, either by rubbing on the skin 
directly, by throwing them on the fire, or through simple procedures such as steam distillation or 
petroleum ether extraction (N. Hill pers. comm.). It is essential that they not irritate the skin, since they 
must be safe and pleasant to use in order to ensure compliance. Although plants with a disagreeable odor 
may be used under conditions of severe mosquito nuisance, those with a pleasant smell will be used more 
often (C. Curtis pers. com.). 

Another important requirement is simplicity of extraction, because production of a highly refined 
botanical compound may prove prohibitively expensive when the yield of bioactive compounds is low. 
For instance, goniothalamin is a mosquito larvicide (LC50 = 5ppm) extracted from Bryonopsis 
laciniosa, but the yield after a multi-step extraction is only 0.45%. 156 Efficacy is also an important 
concern for plant-based repellents, which tend to be very volatile 77 and thus have a shorter duration than 
repellents such as deet. 157 As repellents act in the vapor phase, some active ingredients may be initially 
very effective at repelling mosquitoes, though their effectiveness may rapidly decline as they evaporate. 
Their longevity may be prolonged by incorporating them into oleaginous or semi-solid preparations. 158 

Several field evaluations, where plants were burned to repel mosquitoes, have shown good reduction in 
mosquito landings. 15 ' 20 ' 37 However, smoke has deleterious health effects, 159 and this has to be weighed 
against the potential benefits of repelling mosquitoes by this method. Research needs to be performed to 
discover new topical repellents, such as p-menthane-diol (PMD), that is derived from Corymbia 
citriodora a